
import os
import time
import numpy as np
import pandas as pd
import xgboost as xgb
import matplotlib.pyplot as plt
import seaborn as sns
from xgboost import plot_importance, plot_tree
from sklearn.metrics import mean_squared_error, r2_score, mean_absolute_error

from sklearn.linear_model import LinearRegression, Ridge, Lasso, ElasticNet
from sklearn.svm import SVR
from sklearn.neighbors import KNeighborsRegressor
from sklearn.ensemble import GradientBoostingRegressor, AdaBoostRegressor, RandomForestRegressor
from sklearn.tree import DecisionTreeRegressor
from xgboost import XGBRegressor 
from catboost import CatBoostRegressor


from sklearn.preprocessing import MinMaxScaler, StandardScaler
from sklearn.model_selection import train_test_split, GridSearchCV
pd.set_option('display.max_columns', None)


# Chart drawing
import plotly as py
import plotly.io as pio
import plotly.graph_objects as go
from plotly.subplots import make_subplots
from plotly.offline import download_plotlyjs, init_notebook_mode, plot, iplot

# Mute sklearn warnings
from warnings import simplefilter
simplefilter(action='ignore', category=FutureWarning)
simplefilter(action='ignore', category=DeprecationWarning)

# Show charts when running kernel
#init_notebook_mode(connected=True)

# Change default background color for all visualizations
layout=go.Layout(paper_bgcolor='rgba(0,0,0,0)', plot_bgcolor='rgba(250,250,250,0.8)')
fig = go.Figure(layout=layout)
templated_fig = pio.to_templated(fig)
pio.templates['my_template'] = templated_fig.layout.template
pio.templates.default = 'my_template'

import warnings
warnings.filterwarnings("ignore", category=UserWarning, module="xgboost")
warnings.filterwarnings("ignore")


def evaluate_regression_model(y_true, y_pred):
    """
    Calculate and print evaluation metrics for a regression model.

    Parameters:
    - y_true: Actual values.
    - y_pred: Predicted values.

    Returns:
    - Dictionary containing the evaluation metrics.
    """
    # Calculate evaluation metrics
    mse = mean_squared_error(y_true, y_pred)
    rmse = mean_squared_error(y_true, y_pred, squared=False)
    mae = mean_absolute_error(y_true, y_pred)
    r2 = r2_score(y_true, y_pred)

    # Print the results
    print(f'Mean Squared Error (MSE): {np.round(mse,3)}')
    print(f'Root Mean Squared Error (RMSE): {np.round(rmse,3)}')
    print(f'Mean Absolute Error (MAE): {np.round(mae,3)}')
    print(f'R-squared (R2): {np.round(r2,3)}')

    # Return results as a dictionary
    results = {
        'MSE': mse,
        'RMSE': rmse,
        'MAE': mae,
        'R2': r2
    }

    return results


def evaluate_regression_model2(y_true, y_pred):
    """
    Calculate and print evaluation metrics for a regression model.

    Parameters:
    - y_true: Actual values.
    - y_pred: Predicted values.

    Returns:
    - Dictionary containing the evaluation metrics.
    """
    # Calculate evaluation metrics
    mse = mean_squared_error(y_true, y_pred)
    rmse = mean_squared_error(y_true, y_pred, squared=False)
    mae = mean_absolute_error(y_true, y_pred)
    r2 = r2_score(y_true, y_pred)

#     # Print the results
#     print(f'Mean Squared Error (MSE): {np.round(mse,3)}')
#     print(f'Root Mean Squared Error (RMSE): {np.round(rmse,3)}')
#     print(f'Mean Absolute Error (MAE): {np.round(mae,3)}')
#     print(f'R-squared (R2): {np.round(r2,3)}')

    # Return results as a dictionary
    results = {
        'MSE': mse,
        'RMSE': rmse,
        'MAE': mae,
        'R2': r2
    }

    return results


# Returns RSI values
def rsi(df, periods = 14):
    
    """
    Calculate the Relative Strength Index (RSI) for a given DataFrame.

    Parameters:
    - df (DataFrame): Pandas DataFrame with a 'close' column.
    - periods (int): Number of periods to consider for RSI calculation. Default is 14.

    Returns:
    - Series: A pandas Series containing the RSI values.
    """
        
        
    close = df['close']
    close_delta = close.diff()

    # Make two series: one for lower closes and one for higher closes
    up = close_delta.clip(lower=0)
    down = -1 * close_delta.clip(upper=0)
    
    ma_up = up.ewm(com = periods - 1, adjust=True, min_periods = periods).mean()
    ma_down = down.ewm(com = periods - 1, adjust=True, min_periods = periods).mean()

    rsi = ma_up / ma_down
    rsi = 100 - (100/(1 + rsi))
    return rsi


def gain(x):
    return ((x > 0) * x).sum()


def loss(x):
    return ((x < 0) * x).sum()


def mfi(df, n=14):
    """
    Calculate the Money Flow Index (MFI) for a given DataFrame.

    Parameters:
    - df (DataFrame): Pandas DataFrame with 'high', 'low', 'close', and 'volume' columns.
    - n (int): Number of periods to use for the MFI calculation. Default is 14.

    Returns:
    - numpy.ndarray: An array containing the MFI values.
    """
    
    high = df['high']
    low = df['low']
    close = df['close']
    volume = df['volume']
    
    typical_price = (high + low + close) / 3
    money_flow = typical_price * volume
    mf_sign = np.where(typical_price > typical_price.shift(1), 1, -1)
    signed_mf = money_flow * mf_sign

    # Calculate gain and loss using vectorized operations
    positive_mf = np.where(signed_mf > 0, signed_mf, 0)
    negative_mf = np.where(signed_mf < 0, -signed_mf, 0)

    mf_avg_gain = pd.Series(positive_mf).rolling(n, min_periods=1).sum()
    mf_avg_loss = pd.Series(negative_mf).rolling(n, min_periods=1).sum()

    return (100 - 100 / (1 + mf_avg_gain / mf_avg_loss)).to_numpy()


def plot_regression_accuracy(y_true, y_pred):
    """
    Create various plots to evaluate the accuracy of a linear regression model.

    Parameters:
    - y_true: Actual values.
    - y_pred: Predicted values.
    """
    # Scatter Plot
    plt.scatter(y_true, y_pred)
    plt.xlabel('Actual Values')
    plt.ylabel('Predicted Values')
    plt.title('Scatter Plot of Actual vs Predicted Values')
    plt.show()

    # Residual Plot
    residuals = y_true - y_pred
    plt.scatter(y_pred, residuals)
    plt.axhline(y=0, color='r', linestyle='--')
    plt.xlabel('Predicted Values')
    plt.ylabel('Residuals')
    plt.title('Residual Plot')
    plt.show()

    # Distribution of Residuals
    sns.histplot(residuals, kde=True)
    plt.xlabel('Residuals')
    plt.ylabel('Frequency')
    plt.title('Distribution of Residuals')
    plt.show()

    # Predicted vs Actual Line
    plt.plot(y_true, y_true, linestyle='--', color='r', label='Perfect Fit')
    plt.scatter(y_true, y_pred)
    plt.xlabel('Actual Values')
    plt.ylabel('Predicted Values')
    plt.title('Predicted vs Actual Values with Perfect Fit Line')
    plt.legend()
    plt.show()


def plot_predictions(df,prediction):
    
    """
    Create a Plotly graph to compare actual values with predictions.

    Parameters:
    - df (DataFrame): A pandas DataFrame containing 'date' and 'close_1d_next' columns.
    - prediction (array-like): Predicted values corresponding to the test set.
    """
        
    plot_test_df= df[df.date.dt.year>=2020]
    plot_test_df['prediction'] = prediction

    fig = make_subplots(rows=2, cols=1)
    fig.add_trace(go.Scatter(x=df.date, y=df.close_1d_next,
                             name='Truth',
                             marker_color='LightSkyBlue'), row=1, col=1)

    fig.add_trace(go.Scatter(x=plot_test_df.date,
                             y=plot_test_df.prediction,
                             name='Prediction',
                             marker_color='MediumPurple'), row=1, col=1)
    
    # Add title and Y-axis title for the first subplot
    fig.update_layout(title_text='Train Data and Test Data', title_x=0.5, title_y=0.9)
    fig.update_yaxes(title_text='Prediction', row=1, col=1)

    fig.add_trace(go.Scatter(x=plot_test_df.date,
                             y=y_test,
                             name='Truth',
                             marker_color='LightSkyBlue',
                             showlegend=False), row=2, col=1)

    fig.add_trace(go.Scatter(x=plot_test_df.date,
                             y=prediction,
                             name='Prediction',
                             marker_color='MediumPurple',
                             showlegend=False), row=2, col=1)
    
    fig.update_yaxes(title_text='Prediction', row=2, col=1)

    fig.show()


def plot_feature_importance(model,X_train,top_features):

    """
    Plot the feature importance from a linear regression model and return a sorted DataFrame of feature importances.

    Parameters:
    - model: A trained linear regression model with a coef_ attribute.
    - X_train (DataFrame): The DataFrame used to train the model, for feature names.
    - num_top_features (int): Number of top features to display.

    Returns:
    - DataFrame: Sorted DataFrame with features and their importance.
    """
    
    # Get feature importance scores (coefficients)
    feature_importance = model.coef_

    # Create a DataFrame to store feature names and importance scores
    feature_importance_df = pd.DataFrame({'Feature': X_train.columns, 'Importance': np.abs(feature_importance)})

    # Sort features by importance
    feature_importance_df = feature_importance_df.sort_values(by='Importance', ascending=False).reset_index(drop=True)

    top_features = top_features
    # Plot feature importance
    plt.figure(figsize=(20, 6))
    plt.barh(range(len(feature_importance_df[:top_features])), feature_importance_df[:top_features]['Importance'], align="center")
    plt.yticks(range(len(feature_importance_df[:top_features])), labels=feature_importance_df[:top_features]['Feature'])
    plt.ylabel("Features")
    plt.xlabel("Coefficient Magnitude")
    plt.title(f"Top {top_features} Feature Importance Values")
    plt.show()
    
    return feature_importance_df


out_loc  = '/Users/isapocan/Desktop/LSU/data/'
# Define the file path for the parquet file
parquet_file_path = out_loc + "stock_1d.parquet"


try:
    # Read the Parquet file into a DataFrame
    df = pd.read_parquet(parquet_file_path)

    # Convert column names to lowercase for consistency
    df.columns = df.columns.str.lower()

    # Display the first few rows of the DataFrame
    display(df.head())

except Exception as e:
    print(f"An error occurred while reading the file: {e}")


# Filter the DataFrame to include only rows where 'symbol' is 'MDLZ'
df = df[df['symbol']=='MDLZ']

# Display the first few rows and the shape of the filtered DataFrame
display(df.head())
display(df.shape)

(2733, 15)


def add_moving_averages(df, column_name):
    """
    Adds various moving averages to the DataFrame.

    Parameters:
    - df (DataFrame): The DataFrame to modify.
    - column_name (str): The column name to calculate moving averages for.
    """
    # Exponential Moving Average (EMA)
    df['ema_9'] = df[column_name].ewm(span=9).mean().shift()

    # Simple Moving Averages (SMA) with different periods
    for period in [5, 10, 15, 30]:
        df[f'sma_{period}'] = df[column_name].rolling(window=period).mean().shift()

# Add moving averages for the 'close' column
add_moving_averages(df, 'close')

df.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 2733 entries, 852843 to 855575
Data columns (total 20 columns):
 #   Column                 Non-Null Count  Dtype         
---  ------                 --------------  -----         
 0   date                   2733 non-null   datetime64[ns]
 1   open                   2733 non-null   float64       
 2   high                   2733 non-null   float64       
 3   low                    2733 non-null   float64       
 4   close                  2733 non-null   float64       
 5   adj close              2733 non-null   float64       
 6   volume                 2733 non-null   float64       
 7   symbol                 2733 non-null   object        
 8   security               2733 non-null   object        
 9   gics sector            2733 non-null   object        
 10  gics sub-industry      2733 non-null   object        
 11  headquarters location  2733 non-null   object        
 12  date added             2733 non-null   object        
 13  cik                    2733 non-null   int64         
 14  founded                2733 non-null   object        
 15  ema_9                  2732 non-null   float64       
 16  sma_5                  2728 non-null   float64       
 17  sma_10                 2723 non-null   float64       
 18  sma_15                 2718 non-null   float64       
 19  sma_30                 2703 non-null   float64       
dtypes: datetime64[ns](1), float64(11), int64(1), object(7)
memory usage: 448.4+ KB


df.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 2733 entries, 852843 to 855575
Data columns (total 20 columns):
 #   Column                 Non-Null Count  Dtype         
---  ------                 --------------  -----         
 0   date                   2733 non-null   datetime64[ns]
 1   open                   2733 non-null   float64       
 2   high                   2733 non-null   float64       
 3   low                    2733 non-null   float64       
 4   close                  2733 non-null   float64       
 5   adj close              2733 non-null   float64       
 6   volume                 2733 non-null   float64       
 7   symbol                 2733 non-null   object        
 8   security               2733 non-null   object        
 9   gics sector            2733 non-null   object        
 10  gics sub-industry      2733 non-null   object        
 11  headquarters location  2733 non-null   object        
 12  date added             2733 non-null   object        
 13  cik                    2733 non-null   int64         
 14  founded                2733 non-null   object        
 15  ema_9                  2732 non-null   float64       
 16  sma_5                  2728 non-null   float64       
 17  sma_10                 2723 non-null   float64       
 18  sma_15                 2718 non-null   float64       
 19  sma_30                 2703 non-null   float64       
dtypes: datetime64[ns](1), float64(11), int64(1), object(7)
memory usage: 448.4+ KB


# Add a Relative Strength Index (RSI) column to the DataFrame
try:
    df['rsi'] = rsi(df) # Uncomment and adjust fillna(0) if appropriate for handling missing values
except Exception as e:
    print(f"Error calculating RSI: {e}")

# Add a Money Flow Index (MFI) column to the DataFrame
try:
    df['mfi'] = mfi(df, 14) # The second argument is the period, here assumed to be 14
except Exception as e:
    print(f"Error calculating MFI: {e}")


df[['date','close','ema_9','sma_5','sma_10','sma_15','sma_30','rsi','mfi']]


# Calculate and display the correlation between 'rsi' and 'mfi'
if {'rsi', 'mfi'}.issubset(df.columns):
    correlation = df[['rsi', 'mfi']].corr()
    print(correlation)
else:
    print("DataFrame does not contain 'rsi' and/or 'mfi' columns.")

          rsi       mfi
rsi  1.000000  0.698958
mfi  0.698958  1.000000


# calculating the Moving Average Convergence Divergence (MACD) and its signal line

# Ensure the 'close' column exists in the DataFrame
if 'close' in df.columns:
    # Calculate the 12-period EMA of the closing prices
    df['macd'] = df['close'].ewm(span=12, min_periods=12).mean() - df['close'].ewm(span=26, min_periods=26).mean()

    # Calculate the 9-period EMA of the MACD values (signal line)
    df['macd_signal'] = df['macd'].ewm(span=9, min_periods=9).mean()
else:
    print("DataFrame does not contain 'close' column.")


# Check if the required columns exist in the DataFrame
if {'macd', 'macd_signal'}.issubset(df.columns):
    # Select rows where 'macd' and 'macd_signal' columns do not have missing values
    filtered_df = df[(~df['macd'].isna()) & (~df['macd_signal'].isna())]
    
    # Display the first few rows of these columns
    print(filtered_df[['macd', 'macd_signal']].head())
else:
    print("DataFrame does not contain 'macd' and/or 'macd_signal' columns.")

            macd  macd_signal
852876 -0.147786    -0.050945
852877 -0.175230    -0.078792
852878 -0.198438    -0.104970
852879 -0.235462    -0.132994
852880 -0.226841    -0.152855


# Check if the required columns exist in the DataFrame
if {'date', 'close'}.issubset(df.columns):
    # Create a new column representing the next day's closing price
    df['close_1d_next'] = df['close'].shift(-1)

    # Display the first few rows including 'date', 'close', and 'close_1d_next'
    print(df[['date', 'close', 'close_1d_next']].head())
else:
    print("DataFrame does not contain 'date' and/or 'close' columns.")

             date      close  close_1d_next
852843 2013-01-02  26.670000      26.639999
852844 2013-01-03  26.639999      26.740000
852845 2013-01-04  26.740000      26.660000
852846 2013-01-07  26.660000      26.680000
852847 2013-01-08  26.680000      27.049999


def add_lagged_features(df, column_name, lags):
    """
    Adds lagged features for a specified column in the DataFrame.

    Parameters:
    - df (DataFrame): The DataFrame to modify.
    - column_name (str): The column name to create lagged features for.
    - lags (list of int): The list of lag periods.
    """
    for lag in lags:
        df[f'{column_name}_{lag}d_ago'] = df[column_name].shift(lag)

def add_rolling_avg_features(df, column_name, windows):
    """
    Adds rolling average features for a specified column in the DataFrame.

    Parameters:
    - df (DataFrame): The DataFrame to modify.
    - column_name (str): The column name to create rolling average features for.
    - windows (list of int): The list of rolling window sizes.
    """
    for window in windows:
        df[f'{column_name}_{window}d_avg'] = df[column_name].rolling(window=window).mean()

# Define lag periods and rolling window sizes
lag_periods = [1, 3, 5, 7, 14, 21, 28]
rolling_windows = [3, 5, 7, 10, 15, 30]

# Columns to create features for
columns = ['close', 'adj close', 'open', 'high', 'low', 'volume']

# Add lagged and rolling average features for each column
for column in columns:
    add_lagged_features(df, column, lag_periods)
    add_rolling_avg_features(df, column, rolling_windows)

# View the DataFrame
df.head()


# Check if the DataFrame contains any missing values
if df.isna().any().any():
    # Remove records with missing values and reset the index
    df = df.dropna().reset_index(drop=True)
    print("Missing records removed. DataFrame is now cleaned.")
else:
    df = df.copy()
    print("No missing records found. DataFrame remains unchanged.")

# Display the first few rows of the cleaned DataFrame
df.head()

Missing records removed. DataFrame is now cleaned.


# # Calculate the index for the 70-30 split
# split_index = int(0.7 * len(df))

# # Split the DataFrame into training and testing sets
# train_df = df.iloc[:split_index]
# test_df = df.iloc[split_index:]

# Split the DataFrame into training and testing sets
train_df = df[df.date.dt.year<2020]
test_df = df[df.date.dt.year>=2020]



print(f"Train days: {len(train_df)}, Test days: {len(test_df)}")

fig = go.Figure()
fig.add_trace(go.Scatter(x=train_df.date, y=train_df.close_1d_next, name='Training'))
fig.add_trace(go.Scatter(x=test_df.date,  y=test_df.close_1d_next,  name='Test'))
fig.show()

Train days: 1729, Test days: 970


drop_cols1 = ['date','open','high','low','close','adj close','volume','symbol','security',
 'gics sector','gics sub-industry','headquarters location','date added','cik','founded']

train_df = train_df.drop(drop_cols1, 1)

test_df  = test_df.drop(drop_cols1, 1)

# target column is next day's close price
y_train = train_df['close_1d_next'].copy()
X_train = train_df.drop(['close_1d_next'], 1)

# target column is next day's close price
y_test  = test_df['close_1d_next'].copy()
X_test  = test_df.drop(['close_1d_next'], 1)


scaler = StandardScaler()
X_train_scaled = scaler.fit_transform(X_train)
X_test_scaled = scaler.transform(X_test)


X_train.shape, X_train_scaled.shape, X_test.shape, X_test_scaled.shape,

((1729, 87), (1729, 87), (970, 87), (970, 87))


def train_and_evaluate_models(X_train_scaled,y_train,X_test_scaled,y_test):
    """
    Train and evaluate multiple regression models on a given dataframe.

    Parameters:
    - dataframe: Pandas DataFrame containing the dataset.
    - target_column: Name of the target column (dependent variable).
    - features_columns: List of column names used as features (independent variables).

    Returns:
    - A DataFrame containing evaluation metrics for each model.
    """

    # Split the data into features (X) and target variable (y)
#     X = dataframe[features_columns]
#     y = dataframe[target_column]

    # Split the data into training and testing sets (70-30 split)
    #X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)

    
#     scaler = StandardScaler()
#     X_train_scaled = scaler.fit_transform(X_train)
#     X_test_scaled = scaler.transform(X_test)


    # Initialize the models
    models = {
        'Linear Regression': LinearRegression(),
        'Ridge Regression': Ridge(),
        'Lasso Regression': Lasso(),
        'Elastic Net': ElasticNet(),
        'SVR': SVR(),
        'K-Neighbors Regressor': KNeighborsRegressor(),
        'Decision Tree': DecisionTreeRegressor(),
        'Random Forest': RandomForestRegressor(),
        'Gradient Boosting': GradientBoostingRegressor(),
        'AdaBoost': AdaBoostRegressor(),
        'XGBoost': XGBRegressor(),
        'CatBoost': CatBoostRegressor()
    }

    # Initialize a DataFrame to store the evaluation metrics
    metrics_df = pd.DataFrame(columns=['Model', 'Mean Squared Error', 'Mean Absolute Error', 'R2 Score'])

    # Train and evaluate each model
    for model_name, model in models.items():
        start_time = time.time() 
       
    # Train the model
        model.fit(X_train_scaled, y_train)
        
        end_time = time.time()  # Record the end time

        training_time = end_time - start_time

        # Make predictions
        y_pred = model.predict(X_test_scaled)

        # Evaluate the model
        mse = mean_squared_error(y_test, y_pred)
        mae = mean_absolute_error(y_test, y_pred)
        r2 = r2_score(y_test, y_pred)

        # Store the metrics in the DataFrame
        metrics_df = metrics_df.append({
            'Model': model_name,
            'Mean Squared Error': mse,
            'Mean Absolute Error': mae,
            'R2 Score': r2,
            'Training Time (s)': training_time
        }, ignore_index=True)
        metrics_df = metrics_df.sort_values(by=['R2 Score'],ascending=False)
    return metrics_df


df_compare = train_and_evaluate_models(X_train,y_train,X_test,y_test)

Learning rate set to 0.044643
0:	learn: 6.0603979	total: 63.2ms	remaining: 1m 3s
1:	learn: 5.8169612	total: 69.1ms	remaining: 34.5s
2:	learn: 5.5755845	total: 74ms	remaining: 24.6s
3:	learn: 5.3462380	total: 78.7ms	remaining: 19.6s
4:	learn: 5.1261132	total: 83.9ms	remaining: 16.7s
5:	learn: 4.9208602	total: 88.5ms	remaining: 14.7s
6:	learn: 4.7250714	total: 93.4ms	remaining: 13.2s
7:	learn: 4.5456470	total: 97.7ms	remaining: 12.1s
8:	learn: 4.3606091	total: 102ms	remaining: 11.3s
9:	learn: 4.1830096	total: 107ms	remaining: 10.6s
10:	learn: 4.0166823	total: 111ms	remaining: 10s
11:	learn: 3.8657076	total: 117ms	remaining: 9.62s
12:	learn: 3.7140111	total: 130ms	remaining: 9.84s
13:	learn: 3.5654097	total: 140ms	remaining: 9.83s
14:	learn: 3.4264460	total: 145ms	remaining: 9.54s
15:	learn: 3.2905647	total: 150ms	remaining: 9.22s
16:	learn: 3.1619168	total: 155ms	remaining: 8.95s
17:	learn: 3.0386276	total: 161ms	remaining: 8.76s
18:	learn: 2.9250019	total: 167ms	remaining: 8.61s
19:	learn: 2.8155633	total: 172ms	remaining: 8.42s
20:	learn: 2.7067078	total: 176ms	remaining: 8.2s
21:	learn: 2.6104041	total: 180ms	remaining: 8s
22:	learn: 2.5171443	total: 186ms	remaining: 7.91s
23:	learn: 2.4234694	total: 191ms	remaining: 7.75s
24:	learn: 2.3336920	total: 195ms	remaining: 7.62s
25:	learn: 2.2501270	total: 201ms	remaining: 7.52s
26:	learn: 2.1685983	total: 206ms	remaining: 7.41s
27:	learn: 2.0914990	total: 212ms	remaining: 7.35s
28:	learn: 2.0188179	total: 217ms	remaining: 7.25s
29:	learn: 1.9490958	total: 221ms	remaining: 7.14s
30:	learn: 1.8819748	total: 225ms	remaining: 7.04s
31:	learn: 1.8183164	total: 230ms	remaining: 6.95s
32:	learn: 1.7609789	total: 234ms	remaining: 6.86s
33:	learn: 1.7008286	total: 238ms	remaining: 6.77s
34:	learn: 1.6438044	total: 243ms	remaining: 6.69s
35:	learn: 1.5954289	total: 247ms	remaining: 6.61s
36:	learn: 1.5458456	total: 251ms	remaining: 6.53s
37:	learn: 1.4999957	total: 255ms	remaining: 6.46s
38:	learn: 1.4557749	total: 259ms	remaining: 6.39s
39:	learn: 1.4136122	total: 264ms	remaining: 6.33s
40:	learn: 1.3740245	total: 268ms	remaining: 6.28s
41:	learn: 1.3360847	total: 273ms	remaining: 6.21s
42:	learn: 1.3006229	total: 277ms	remaining: 6.16s
43:	learn: 1.2683312	total: 281ms	remaining: 6.11s
44:	learn: 1.2328651	total: 285ms	remaining: 6.05s
45:	learn: 1.2022101	total: 289ms	remaining: 6s
46:	learn: 1.1716981	total: 294ms	remaining: 5.95s
47:	learn: 1.1425806	total: 298ms	remaining: 5.91s
48:	learn: 1.1126613	total: 302ms	remaining: 5.86s
49:	learn: 1.0845503	total: 306ms	remaining: 5.82s
50:	learn: 1.0590177	total: 311ms	remaining: 5.78s
51:	learn: 1.0347707	total: 315ms	remaining: 5.74s
52:	learn: 1.0113231	total: 319ms	remaining: 5.7s
53:	learn: 0.9886945	total: 323ms	remaining: 5.66s
54:	learn: 0.9669580	total: 328ms	remaining: 5.63s
55:	learn: 0.9468328	total: 332ms	remaining: 5.59s
56:	learn: 0.9270737	total: 336ms	remaining: 5.56s
57:	learn: 0.9100334	total: 340ms	remaining: 5.52s
58:	learn: 0.8917612	total: 345ms	remaining: 5.5s
59:	learn: 0.8750815	total: 349ms	remaining: 5.47s
60:	learn: 0.8601060	total: 353ms	remaining: 5.44s
61:	learn: 0.8462298	total: 358ms	remaining: 5.41s
62:	learn: 0.8345633	total: 362ms	remaining: 5.38s
63:	learn: 0.8216135	total: 367ms	remaining: 5.36s
64:	learn: 0.8083116	total: 371ms	remaining: 5.34s
65:	learn: 0.7956239	total: 375ms	remaining: 5.31s
66:	learn: 0.7843834	total: 380ms	remaining: 5.29s
67:	learn: 0.7740919	total: 384ms	remaining: 5.26s
68:	learn: 0.7643108	total: 388ms	remaining: 5.24s
69:	learn: 0.7561664	total: 393ms	remaining: 5.22s
70:	learn: 0.7491327	total: 397ms	remaining: 5.2s
71:	learn: 0.7411352	total: 403ms	remaining: 5.2s
72:	learn: 0.7330505	total: 408ms	remaining: 5.18s
73:	learn: 0.7269962	total: 413ms	remaining: 5.17s
74:	learn: 0.7192950	total: 418ms	remaining: 5.15s
75:	learn: 0.7126213	total: 422ms	remaining: 5.13s
76:	learn: 0.7067352	total: 426ms	remaining: 5.11s
77:	learn: 0.7006922	total: 430ms	remaining: 5.08s
78:	learn: 0.6944183	total: 435ms	remaining: 5.07s
79:	learn: 0.6885217	total: 439ms	remaining: 5.05s
80:	learn: 0.6825927	total: 443ms	remaining: 5.03s
81:	learn: 0.6783627	total: 448ms	remaining: 5.01s
82:	learn: 0.6741644	total: 452ms	remaining: 5s
83:	learn: 0.6696567	total: 456ms	remaining: 4.98s
84:	learn: 0.6646154	total: 461ms	remaining: 4.96s
85:	learn: 0.6599463	total: 465ms	remaining: 4.95s
86:	learn: 0.6560602	total: 470ms	remaining: 4.93s
87:	learn: 0.6536929	total: 474ms	remaining: 4.91s
88:	learn: 0.6496752	total: 478ms	remaining: 4.89s
89:	learn: 0.6462499	total: 483ms	remaining: 4.89s
90:	learn: 0.6428735	total: 488ms	remaining: 4.88s
91:	learn: 0.6397495	total: 494ms	remaining: 4.87s
92:	learn: 0.6367543	total: 499ms	remaining: 4.86s
93:	learn: 0.6337726	total: 503ms	remaining: 4.84s
94:	learn: 0.6308954	total: 507ms	remaining: 4.83s
95:	learn: 0.6280612	total: 511ms	remaining: 4.81s
96:	learn: 0.6250764	total: 515ms	remaining: 4.79s
97:	learn: 0.6223572	total: 519ms	remaining: 4.78s
98:	learn: 0.6205398	total: 524ms	remaining: 4.77s
99:	learn: 0.6186628	total: 528ms	remaining: 4.75s
100:	learn: 0.6154184	total: 532ms	remaining: 4.74s
101:	learn: 0.6124974	total: 536ms	remaining: 4.72s
102:	learn: 0.6095369	total: 541ms	remaining: 4.71s
103:	learn: 0.6081161	total: 545ms	remaining: 4.69s
104:	learn: 0.6061648	total: 549ms	remaining: 4.68s
105:	learn: 0.6037361	total: 554ms	remaining: 4.67s
106:	learn: 0.6018697	total: 558ms	remaining: 4.66s
107:	learn: 0.5996563	total: 562ms	remaining: 4.64s
108:	learn: 0.5987782	total: 567ms	remaining: 4.63s
109:	learn: 0.5962953	total: 571ms	remaining: 4.62s
110:	learn: 0.5944525	total: 575ms	remaining: 4.61s
111:	learn: 0.5925792	total: 580ms	remaining: 4.59s
112:	learn: 0.5908725	total: 584ms	remaining: 4.59s
113:	learn: 0.5892615	total: 590ms	remaining: 4.58s
114:	learn: 0.5876082	total: 595ms	remaining: 4.58s
115:	learn: 0.5856574	total: 600ms	remaining: 4.58s
116:	learn: 0.5839197	total: 606ms	remaining: 4.57s
117:	learn: 0.5821181	total: 611ms	remaining: 4.57s
118:	learn: 0.5802687	total: 617ms	remaining: 4.57s
119:	learn: 0.5783092	total: 621ms	remaining: 4.55s
120:	learn: 0.5768204	total: 625ms	remaining: 4.54s
121:	learn: 0.5753784	total: 629ms	remaining: 4.53s
122:	learn: 0.5739414	total: 634ms	remaining: 4.52s
123:	learn: 0.5731172	total: 638ms	remaining: 4.51s
124:	learn: 0.5717661	total: 642ms	remaining: 4.5s
125:	learn: 0.5699490	total: 647ms	remaining: 4.49s
126:	learn: 0.5692480	total: 651ms	remaining: 4.47s
127:	learn: 0.5683714	total: 655ms	remaining: 4.46s
128:	learn: 0.5669671	total: 659ms	remaining: 4.45s
129:	learn: 0.5661305	total: 664ms	remaining: 4.44s
130:	learn: 0.5648214	total: 668ms	remaining: 4.43s
131:	learn: 0.5632118	total: 672ms	remaining: 4.42s
132:	learn: 0.5621830	total: 676ms	remaining: 4.41s
133:	learn: 0.5614073	total: 680ms	remaining: 4.4s
134:	learn: 0.5607392	total: 685ms	remaining: 4.39s
135:	learn: 0.5592399	total: 689ms	remaining: 4.38s
136:	learn: 0.5582825	total: 694ms	remaining: 4.37s
137:	learn: 0.5570203	total: 698ms	remaining: 4.36s
138:	learn: 0.5555467	total: 702ms	remaining: 4.35s
139:	learn: 0.5544866	total: 707ms	remaining: 4.34s
140:	learn: 0.5534618	total: 712ms	remaining: 4.34s
141:	learn: 0.5527745	total: 716ms	remaining: 4.33s
142:	learn: 0.5517276	total: 720ms	remaining: 4.32s
143:	learn: 0.5511754	total: 725ms	remaining: 4.31s
144:	learn: 0.5499848	total: 729ms	remaining: 4.3s
145:	learn: 0.5492370	total: 733ms	remaining: 4.29s
146:	learn: 0.5480831	total: 737ms	remaining: 4.28s
147:	learn: 0.5470207	total: 741ms	remaining: 4.26s
148:	learn: 0.5461768	total: 745ms	remaining: 4.26s
149:	learn: 0.5452187	total: 750ms	remaining: 4.25s
150:	learn: 0.5442155	total: 754ms	remaining: 4.24s
151:	learn: 0.5431025	total: 758ms	remaining: 4.23s
152:	learn: 0.5423642	total: 762ms	remaining: 4.22s
153:	learn: 0.5414228	total: 767ms	remaining: 4.21s
154:	learn: 0.5402482	total: 771ms	remaining: 4.2s
155:	learn: 0.5390607	total: 775ms	remaining: 4.19s
156:	learn: 0.5383473	total: 779ms	remaining: 4.18s
157:	learn: 0.5373131	total: 784ms	remaining: 4.18s
158:	learn: 0.5364392	total: 789ms	remaining: 4.17s
159:	learn: 0.5357421	total: 793ms	remaining: 4.16s
160:	learn: 0.5343277	total: 798ms	remaining: 4.16s
161:	learn: 0.5337785	total: 803ms	remaining: 4.16s
162:	learn: 0.5329012	total: 809ms	remaining: 4.15s
163:	learn: 0.5317662	total: 814ms	remaining: 4.15s
164:	learn: 0.5305788	total: 819ms	remaining: 4.14s
165:	learn: 0.5291670	total: 823ms	remaining: 4.13s
166:	learn: 0.5279711	total: 827ms	remaining: 4.13s
167:	learn: 0.5263201	total: 831ms	remaining: 4.12s
168:	learn: 0.5250413	total: 836ms	remaining: 4.11s
169:	learn: 0.5239617	total: 840ms	remaining: 4.1s
170:	learn: 0.5232086	total: 844ms	remaining: 4.09s
171:	learn: 0.5223393	total: 849ms	remaining: 4.09s
172:	learn: 0.5213785	total: 853ms	remaining: 4.08s
173:	learn: 0.5205693	total: 858ms	remaining: 4.07s
174:	learn: 0.5195080	total: 862ms	remaining: 4.06s
175:	learn: 0.5184230	total: 866ms	remaining: 4.05s
176:	learn: 0.5179592	total: 870ms	remaining: 4.05s
177:	learn: 0.5170185	total: 874ms	remaining: 4.04s
178:	learn: 0.5158612	total: 879ms	remaining: 4.03s
179:	learn: 0.5145391	total: 883ms	remaining: 4.02s
180:	learn: 0.5134332	total: 887ms	remaining: 4.01s
181:	learn: 0.5123862	total: 891ms	remaining: 4.01s
182:	learn: 0.5109381	total: 896ms	remaining: 4s
183:	learn: 0.5094690	total: 900ms	remaining: 3.99s
184:	learn: 0.5087636	total: 904ms	remaining: 3.98s
185:	learn: 0.5082193	total: 908ms	remaining: 3.97s
186:	learn: 0.5077741	total: 912ms	remaining: 3.97s
187:	learn: 0.5064784	total: 917ms	remaining: 3.96s
188:	learn: 0.5053126	total: 921ms	remaining: 3.95s
189:	learn: 0.5045861	total: 925ms	remaining: 3.94s
190:	learn: 0.5035784	total: 929ms	remaining: 3.94s
191:	learn: 0.5025377	total: 934ms	remaining: 3.93s
192:	learn: 0.5017748	total: 938ms	remaining: 3.92s
193:	learn: 0.5008812	total: 942ms	remaining: 3.91s
194:	learn: 0.5004256	total: 946ms	remaining: 3.91s
195:	learn: 0.4994655	total: 951ms	remaining: 3.9s
196:	learn: 0.4989930	total: 955ms	remaining: 3.89s
197:	learn: 0.4980951	total: 959ms	remaining: 3.88s
198:	learn: 0.4973099	total: 963ms	remaining: 3.88s
199:	learn: 0.4961640	total: 968ms	remaining: 3.87s
200:	learn: 0.4946290	total: 972ms	remaining: 3.86s
201:	learn: 0.4940315	total: 976ms	remaining: 3.85s
202:	learn: 0.4931395	total: 981ms	remaining: 3.85s
203:	learn: 0.4920884	total: 985ms	remaining: 3.85s
204:	learn: 0.4909474	total: 990ms	remaining: 3.84s
205:	learn: 0.4899796	total: 996ms	remaining: 3.84s
206:	learn: 0.4890645	total: 1s	remaining: 3.83s
207:	learn: 0.4882374	total: 1s	remaining: 3.82s
208:	learn: 0.4871172	total: 1.01s	remaining: 3.82s
209:	learn: 0.4861811	total: 1.01s	remaining: 3.81s
210:	learn: 0.4850126	total: 1.02s	remaining: 3.8s
211:	learn: 0.4837164	total: 1.02s	remaining: 3.8s
212:	learn: 0.4829656	total: 1.02s	remaining: 3.79s
213:	learn: 0.4819662	total: 1.03s	remaining: 3.78s
214:	learn: 0.4817394	total: 1.03s	remaining: 3.77s
215:	learn: 0.4811177	total: 1.04s	remaining: 3.77s
216:	learn: 0.4806224	total: 1.04s	remaining: 3.76s
217:	learn: 0.4800808	total: 1.05s	remaining: 3.75s
218:	learn: 0.4789126	total: 1.05s	remaining: 3.75s
219:	learn: 0.4784037	total: 1.05s	remaining: 3.74s
220:	learn: 0.4774369	total: 1.06s	remaining: 3.74s
221:	learn: 0.4764525	total: 1.06s	remaining: 3.73s
222:	learn: 0.4757860	total: 1.07s	remaining: 3.72s
223:	learn: 0.4748362	total: 1.07s	remaining: 3.72s
224:	learn: 0.4738011	total: 1.08s	remaining: 3.71s
225:	learn: 0.4732533	total: 1.08s	remaining: 3.71s
226:	learn: 0.4724302	total: 1.09s	remaining: 3.7s
227:	learn: 0.4714185	total: 1.09s	remaining: 3.7s
228:	learn: 0.4704901	total: 1.1s	remaining: 3.69s
229:	learn: 0.4696393	total: 1.1s	remaining: 3.68s
230:	learn: 0.4688642	total: 1.1s	remaining: 3.68s
231:	learn: 0.4684056	total: 1.11s	remaining: 3.67s
232:	learn: 0.4680263	total: 1.11s	remaining: 3.66s
233:	learn: 0.4673195	total: 1.12s	remaining: 3.66s
234:	learn: 0.4661429	total: 1.12s	remaining: 3.65s
235:	learn: 0.4651517	total: 1.13s	remaining: 3.65s
236:	learn: 0.4643784	total: 1.13s	remaining: 3.65s
237:	learn: 0.4639422	total: 1.14s	remaining: 3.64s
238:	learn: 0.4629527	total: 1.14s	remaining: 3.63s
239:	learn: 0.4621494	total: 1.15s	remaining: 3.63s
240:	learn: 0.4611603	total: 1.15s	remaining: 3.62s
241:	learn: 0.4603838	total: 1.15s	remaining: 3.62s
242:	learn: 0.4593452	total: 1.16s	remaining: 3.61s
243:	learn: 0.4586434	total: 1.16s	remaining: 3.6s
244:	learn: 0.4577788	total: 1.17s	remaining: 3.6s
245:	learn: 0.4576161	total: 1.17s	remaining: 3.59s
246:	learn: 0.4570526	total: 1.18s	remaining: 3.59s
247:	learn: 0.4560632	total: 1.18s	remaining: 3.59s
248:	learn: 0.4552049	total: 1.19s	remaining: 3.59s
249:	learn: 0.4544847	total: 1.2s	remaining: 3.59s
250:	learn: 0.4536628	total: 1.2s	remaining: 3.58s
251:	learn: 0.4529050	total: 1.21s	remaining: 3.58s
252:	learn: 0.4521824	total: 1.21s	remaining: 3.57s
253:	learn: 0.4515072	total: 1.21s	remaining: 3.57s
254:	learn: 0.4506698	total: 1.22s	remaining: 3.56s
255:	learn: 0.4501205	total: 1.22s	remaining: 3.56s
256:	learn: 0.4493848	total: 1.23s	remaining: 3.55s
257:	learn: 0.4486159	total: 1.23s	remaining: 3.55s
258:	learn: 0.4482442	total: 1.24s	remaining: 3.54s
259:	learn: 0.4478827	total: 1.24s	remaining: 3.54s
260:	learn: 0.4473978	total: 1.25s	remaining: 3.53s
261:	learn: 0.4468994	total: 1.25s	remaining: 3.53s
262:	learn: 0.4456701	total: 1.26s	remaining: 3.52s
263:	learn: 0.4449110	total: 1.26s	remaining: 3.52s
264:	learn: 0.4442542	total: 1.27s	remaining: 3.51s
265:	learn: 0.4431941	total: 1.27s	remaining: 3.51s
266:	learn: 0.4425103	total: 1.27s	remaining: 3.5s
267:	learn: 0.4416661	total: 1.28s	remaining: 3.49s
268:	learn: 0.4411630	total: 1.28s	remaining: 3.49s
269:	learn: 0.4404441	total: 1.29s	remaining: 3.48s
270:	learn: 0.4400363	total: 1.29s	remaining: 3.48s
271:	learn: 0.4398475	total: 1.3s	remaining: 3.47s
272:	learn: 0.4390363	total: 1.3s	remaining: 3.47s
273:	learn: 0.4381161	total: 1.31s	remaining: 3.46s
274:	learn: 0.4374399	total: 1.31s	remaining: 3.46s
275:	learn: 0.4370343	total: 1.31s	remaining: 3.45s
276:	learn: 0.4366440	total: 1.32s	remaining: 3.45s
277:	learn: 0.4363497	total: 1.32s	remaining: 3.44s
278:	learn: 0.4360749	total: 1.33s	remaining: 3.43s
279:	learn: 0.4353030	total: 1.33s	remaining: 3.43s
280:	learn: 0.4346082	total: 1.34s	remaining: 3.42s
281:	learn: 0.4342305	total: 1.34s	remaining: 3.42s
282:	learn: 0.4338120	total: 1.34s	remaining: 3.41s
283:	learn: 0.4336984	total: 1.35s	remaining: 3.4s
284:	learn: 0.4329844	total: 1.35s	remaining: 3.4s
285:	learn: 0.4325585	total: 1.36s	remaining: 3.39s
286:	learn: 0.4316236	total: 1.36s	remaining: 3.38s
287:	learn: 0.4312618	total: 1.37s	remaining: 3.38s
288:	learn: 0.4305788	total: 1.37s	remaining: 3.37s
289:	learn: 0.4299323	total: 1.38s	remaining: 3.37s
290:	learn: 0.4294147	total: 1.38s	remaining: 3.37s
291:	learn: 0.4287311	total: 1.39s	remaining: 3.36s
292:	learn: 0.4281021	total: 1.39s	remaining: 3.35s
293:	learn: 0.4274824	total: 1.39s	remaining: 3.35s
294:	learn: 0.4266494	total: 1.4s	remaining: 3.34s
295:	learn: 0.4259133	total: 1.4s	remaining: 3.34s
296:	learn: 0.4254208	total: 1.41s	remaining: 3.33s
297:	learn: 0.4248581	total: 1.41s	remaining: 3.33s
298:	learn: 0.4244197	total: 1.42s	remaining: 3.33s
299:	learn: 0.4238834	total: 1.42s	remaining: 3.32s
300:	learn: 0.4235753	total: 1.43s	remaining: 3.31s
301:	learn: 0.4230278	total: 1.43s	remaining: 3.31s
302:	learn: 0.4223168	total: 1.44s	remaining: 3.3s
303:	learn: 0.4218202	total: 1.44s	remaining: 3.3s
304:	learn: 0.4213324	total: 1.44s	remaining: 3.29s
305:	learn: 0.4209713	total: 1.45s	remaining: 3.29s
306:	learn: 0.4206682	total: 1.45s	remaining: 3.28s
307:	learn: 0.4205730	total: 1.46s	remaining: 3.27s
308:	learn: 0.4200366	total: 1.46s	remaining: 3.27s
309:	learn: 0.4196622	total: 1.47s	remaining: 3.26s
310:	learn: 0.4188254	total: 1.47s	remaining: 3.25s
311:	learn: 0.4182222	total: 1.47s	remaining: 3.25s
312:	learn: 0.4181272	total: 1.48s	remaining: 3.24s
313:	learn: 0.4178863	total: 1.48s	remaining: 3.24s
314:	learn: 0.4171997	total: 1.49s	remaining: 3.23s
315:	learn: 0.4165719	total: 1.49s	remaining: 3.23s
316:	learn: 0.4158275	total: 1.49s	remaining: 3.22s
317:	learn: 0.4150223	total: 1.5s	remaining: 3.21s
318:	learn: 0.4144227	total: 1.5s	remaining: 3.21s
319:	learn: 0.4138124	total: 1.51s	remaining: 3.2s
320:	learn: 0.4135332	total: 1.51s	remaining: 3.2s
321:	learn: 0.4129018	total: 1.51s	remaining: 3.19s
322:	learn: 0.4123695	total: 1.52s	remaining: 3.18s
323:	learn: 0.4116624	total: 1.52s	remaining: 3.18s
324:	learn: 0.4108634	total: 1.53s	remaining: 3.17s
325:	learn: 0.4103299	total: 1.53s	remaining: 3.17s
326:	learn: 0.4096449	total: 1.54s	remaining: 3.16s
327:	learn: 0.4092467	total: 1.54s	remaining: 3.16s
328:	learn: 0.4090363	total: 1.54s	remaining: 3.15s
329:	learn: 0.4080857	total: 1.55s	remaining: 3.14s
330:	learn: 0.4077160	total: 1.55s	remaining: 3.14s
331:	learn: 0.4071226	total: 1.56s	remaining: 3.13s
332:	learn: 0.4067749	total: 1.56s	remaining: 3.13s
333:	learn: 0.4062413	total: 1.57s	remaining: 3.12s
334:	learn: 0.4055695	total: 1.57s	remaining: 3.12s
335:	learn: 0.4054942	total: 1.57s	remaining: 3.11s
336:	learn: 0.4047835	total: 1.58s	remaining: 3.11s
337:	learn: 0.4042196	total: 1.58s	remaining: 3.1s
338:	learn: 0.4038735	total: 1.59s	remaining: 3.1s
339:	learn: 0.4036386	total: 1.59s	remaining: 3.1s
340:	learn: 0.4033409	total: 1.6s	remaining: 3.09s
341:	learn: 0.4026643	total: 1.6s	remaining: 3.08s
342:	learn: 0.4023055	total: 1.61s	remaining: 3.08s
343:	learn: 0.4021658	total: 1.61s	remaining: 3.07s
344:	learn: 0.4018412	total: 1.62s	remaining: 3.07s
345:	learn: 0.4013429	total: 1.62s	remaining: 3.06s
346:	learn: 0.4008246	total: 1.63s	remaining: 3.06s
347:	learn: 0.4002585	total: 1.63s	remaining: 3.05s
348:	learn: 0.3998110	total: 1.63s	remaining: 3.05s
349:	learn: 0.3990401	total: 1.64s	remaining: 3.04s
350:	learn: 0.3982868	total: 1.64s	remaining: 3.04s
351:	learn: 0.3977941	total: 1.65s	remaining: 3.03s
352:	learn: 0.3971540	total: 1.65s	remaining: 3.03s
353:	learn: 0.3966626	total: 1.66s	remaining: 3.02s
354:	learn: 0.3965989	total: 1.66s	remaining: 3.01s
355:	learn: 0.3962189	total: 1.66s	remaining: 3.01s
356:	learn: 0.3958068	total: 1.67s	remaining: 3s
357:	learn: 0.3953095	total: 1.67s	remaining: 3s
358:	learn: 0.3948418	total: 1.68s	remaining: 2.99s
359:	learn: 0.3946123	total: 1.68s	remaining: 2.99s
360:	learn: 0.3938100	total: 1.69s	remaining: 2.98s
361:	learn: 0.3932202	total: 1.69s	remaining: 2.98s
362:	learn: 0.3925485	total: 1.69s	remaining: 2.97s
363:	learn: 0.3918346	total: 1.7s	remaining: 2.97s
364:	learn: 0.3915180	total: 1.7s	remaining: 2.96s
365:	learn: 0.3907133	total: 1.71s	remaining: 2.96s
366:	learn: 0.3905787	total: 1.71s	remaining: 2.95s
367:	learn: 0.3900100	total: 1.72s	remaining: 2.95s
368:	learn: 0.3894581	total: 1.72s	remaining: 2.94s
369:	learn: 0.3887745	total: 1.73s	remaining: 2.94s
370:	learn: 0.3881815	total: 1.73s	remaining: 2.93s
371:	learn: 0.3877839	total: 1.73s	remaining: 2.93s
372:	learn: 0.3872649	total: 1.74s	remaining: 2.92s
373:	learn: 0.3864811	total: 1.74s	remaining: 2.92s
374:	learn: 0.3859586	total: 1.75s	remaining: 2.91s
375:	learn: 0.3857552	total: 1.75s	remaining: 2.9s
376:	learn: 0.3851249	total: 1.75s	remaining: 2.9s
377:	learn: 0.3846241	total: 1.76s	remaining: 2.9s
378:	learn: 0.3842221	total: 1.76s	remaining: 2.89s
379:	learn: 0.3840501	total: 1.77s	remaining: 2.88s
380:	learn: 0.3839565	total: 1.77s	remaining: 2.88s
381:	learn: 0.3834811	total: 1.78s	remaining: 2.88s
382:	learn: 0.3829302	total: 1.78s	remaining: 2.87s
383:	learn: 0.3824233	total: 1.79s	remaining: 2.87s
384:	learn: 0.3820862	total: 1.79s	remaining: 2.86s
385:	learn: 0.3817059	total: 1.79s	remaining: 2.86s
386:	learn: 0.3810940	total: 1.8s	remaining: 2.85s
387:	learn: 0.3807112	total: 1.8s	remaining: 2.85s
388:	learn: 0.3802605	total: 1.81s	remaining: 2.84s
389:	learn: 0.3798039	total: 1.81s	remaining: 2.83s
390:	learn: 0.3793238	total: 1.82s	remaining: 2.83s
391:	learn: 0.3787842	total: 1.82s	remaining: 2.83s
392:	learn: 0.3783469	total: 1.82s	remaining: 2.82s
393:	learn: 0.3774757	total: 1.83s	remaining: 2.81s
394:	learn: 0.3770911	total: 1.83s	remaining: 2.81s
395:	learn: 0.3767255	total: 1.84s	remaining: 2.8s
396:	learn: 0.3761264	total: 1.84s	remaining: 2.8s
397:	learn: 0.3757592	total: 1.85s	remaining: 2.79s
398:	learn: 0.3751453	total: 1.85s	remaining: 2.79s
399:	learn: 0.3748033	total: 1.85s	remaining: 2.78s
400:	learn: 0.3743184	total: 1.86s	remaining: 2.78s
401:	learn: 0.3740077	total: 1.86s	remaining: 2.77s
402:	learn: 0.3733825	total: 1.87s	remaining: 2.77s
403:	learn: 0.3728581	total: 1.87s	remaining: 2.76s
404:	learn: 0.3722381	total: 1.88s	remaining: 2.76s
405:	learn: 0.3716600	total: 1.88s	remaining: 2.75s
406:	learn: 0.3709735	total: 1.89s	remaining: 2.75s
407:	learn: 0.3706292	total: 1.89s	remaining: 2.74s
408:	learn: 0.3700405	total: 1.89s	remaining: 2.74s
409:	learn: 0.3695724	total: 1.9s	remaining: 2.73s
410:	learn: 0.3688604	total: 1.9s	remaining: 2.73s
411:	learn: 0.3687772	total: 1.91s	remaining: 2.72s
412:	learn: 0.3680599	total: 1.91s	remaining: 2.72s
413:	learn: 0.3676867	total: 1.92s	remaining: 2.71s
414:	learn: 0.3673532	total: 1.92s	remaining: 2.71s
415:	learn: 0.3667493	total: 1.92s	remaining: 2.7s
416:	learn: 0.3663581	total: 1.93s	remaining: 2.69s
417:	learn: 0.3659273	total: 1.93s	remaining: 2.69s
418:	learn: 0.3653585	total: 1.94s	remaining: 2.69s
419:	learn: 0.3647104	total: 1.94s	remaining: 2.68s
420:	learn: 0.3643467	total: 1.94s	remaining: 2.67s
421:	learn: 0.3638706	total: 1.95s	remaining: 2.67s
422:	learn: 0.3637044	total: 1.95s	remaining: 2.66s
423:	learn: 0.3630532	total: 1.96s	remaining: 2.66s
424:	learn: 0.3628976	total: 1.96s	remaining: 2.65s
425:	learn: 0.3622257	total: 1.97s	remaining: 2.65s
426:	learn: 0.3618914	total: 1.97s	remaining: 2.65s
427:	learn: 0.3614434	total: 1.98s	remaining: 2.64s
428:	learn: 0.3610056	total: 1.98s	remaining: 2.64s
429:	learn: 0.3603069	total: 1.99s	remaining: 2.63s
430:	learn: 0.3599433	total: 1.99s	remaining: 2.63s
431:	learn: 0.3598531	total: 1.99s	remaining: 2.62s
432:	learn: 0.3597998	total: 2s	remaining: 2.62s
433:	learn: 0.3592574	total: 2s	remaining: 2.61s
434:	learn: 0.3586937	total: 2.01s	remaining: 2.6s
435:	learn: 0.3582985	total: 2.01s	remaining: 2.6s
436:	learn: 0.3579804	total: 2.01s	remaining: 2.6s
437:	learn: 0.3576664	total: 2.02s	remaining: 2.59s
438:	learn: 0.3572953	total: 2.02s	remaining: 2.58s
439:	learn: 0.3568969	total: 2.03s	remaining: 2.58s
440:	learn: 0.3562306	total: 2.03s	remaining: 2.58s
441:	learn: 0.3555624	total: 2.04s	remaining: 2.57s
442:	learn: 0.3550610	total: 2.04s	remaining: 2.56s
443:	learn: 0.3545855	total: 2.04s	remaining: 2.56s
444:	learn: 0.3541245	total: 2.04s	remaining: 2.55s
445:	learn: 0.3537684	total: 2.05s	remaining: 2.54s
446:	learn: 0.3534726	total: 2.05s	remaining: 2.54s
447:	learn: 0.3533520	total: 2.06s	remaining: 2.53s
448:	learn: 0.3530596	total: 2.06s	remaining: 2.53s
449:	learn: 0.3525138	total: 2.06s	remaining: 2.52s
450:	learn: 0.3520520	total: 2.07s	remaining: 2.52s
451:	learn: 0.3517294	total: 2.07s	remaining: 2.51s
452:	learn: 0.3516567	total: 2.08s	remaining: 2.5s
453:	learn: 0.3513043	total: 2.08s	remaining: 2.5s
454:	learn: 0.3509406	total: 2.08s	remaining: 2.49s
455:	learn: 0.3504919	total: 2.08s	remaining: 2.49s
456:	learn: 0.3498914	total: 2.09s	remaining: 2.48s
457:	learn: 0.3494417	total: 2.09s	remaining: 2.48s
458:	learn: 0.3490944	total: 2.1s	remaining: 2.47s
459:	learn: 0.3487102	total: 2.1s	remaining: 2.46s
460:	learn: 0.3484557	total: 2.1s	remaining: 2.46s
461:	learn: 0.3478789	total: 2.11s	remaining: 2.46s
462:	learn: 0.3474275	total: 2.11s	remaining: 2.45s
463:	learn: 0.3468342	total: 2.12s	remaining: 2.45s
464:	learn: 0.3463040	total: 2.12s	remaining: 2.44s
465:	learn: 0.3458213	total: 2.13s	remaining: 2.44s
466:	learn: 0.3451803	total: 2.13s	remaining: 2.43s
467:	learn: 0.3447292	total: 2.14s	remaining: 2.43s
468:	learn: 0.3443530	total: 2.14s	remaining: 2.42s
469:	learn: 0.3437850	total: 2.15s	remaining: 2.42s
470:	learn: 0.3433565	total: 2.15s	remaining: 2.41s
471:	learn: 0.3430048	total: 2.15s	remaining: 2.41s
472:	learn: 0.3425770	total: 2.16s	remaining: 2.4s
473:	learn: 0.3419536	total: 2.16s	remaining: 2.4s
474:	learn: 0.3414171	total: 2.17s	remaining: 2.4s
475:	learn: 0.3409048	total: 2.17s	remaining: 2.39s
476:	learn: 0.3406389	total: 2.18s	remaining: 2.39s
477:	learn: 0.3402231	total: 2.18s	remaining: 2.38s
478:	learn: 0.3397786	total: 2.19s	remaining: 2.38s
479:	learn: 0.3396398	total: 2.19s	remaining: 2.37s
480:	learn: 0.3395825	total: 2.19s	remaining: 2.37s
481:	learn: 0.3391371	total: 2.2s	remaining: 2.36s
482:	learn: 0.3386234	total: 2.2s	remaining: 2.36s
483:	learn: 0.3382257	total: 2.21s	remaining: 2.35s
484:	learn: 0.3378495	total: 2.21s	remaining: 2.35s
485:	learn: 0.3375195	total: 2.22s	remaining: 2.34s
486:	learn: 0.3371611	total: 2.22s	remaining: 2.34s
487:	learn: 0.3367422	total: 2.22s	remaining: 2.33s
488:	learn: 0.3363054	total: 2.23s	remaining: 2.33s
489:	learn: 0.3359786	total: 2.23s	remaining: 2.32s
490:	learn: 0.3356998	total: 2.24s	remaining: 2.32s
491:	learn: 0.3354453	total: 2.24s	remaining: 2.31s
492:	learn: 0.3352187	total: 2.25s	remaining: 2.31s
493:	learn: 0.3350124	total: 2.25s	remaining: 2.31s
494:	learn: 0.3346929	total: 2.25s	remaining: 2.3s
495:	learn: 0.3342032	total: 2.26s	remaining: 2.3s
496:	learn: 0.3336972	total: 2.26s	remaining: 2.29s
497:	learn: 0.3336357	total: 2.27s	remaining: 2.29s
498:	learn: 0.3333691	total: 2.27s	remaining: 2.28s
499:	learn: 0.3329063	total: 2.28s	remaining: 2.28s
500:	learn: 0.3326459	total: 2.28s	remaining: 2.27s
501:	learn: 0.3323511	total: 2.28s	remaining: 2.27s
502:	learn: 0.3319477	total: 2.29s	remaining: 2.26s
503:	learn: 0.3317364	total: 2.29s	remaining: 2.26s
504:	learn: 0.3312039	total: 2.3s	remaining: 2.25s
505:	learn: 0.3305307	total: 2.3s	remaining: 2.25s
506:	learn: 0.3300698	total: 2.31s	remaining: 2.24s
507:	learn: 0.3296915	total: 2.31s	remaining: 2.24s
508:	learn: 0.3294125	total: 2.31s	remaining: 2.23s
509:	learn: 0.3291011	total: 2.32s	remaining: 2.23s
510:	learn: 0.3289021	total: 2.32s	remaining: 2.22s
511:	learn: 0.3285137	total: 2.33s	remaining: 2.22s
512:	learn: 0.3278964	total: 2.33s	remaining: 2.21s
513:	learn: 0.3274555	total: 2.34s	remaining: 2.21s
514:	learn: 0.3270691	total: 2.34s	remaining: 2.2s
515:	learn: 0.3268052	total: 2.35s	remaining: 2.2s
516:	learn: 0.3263295	total: 2.35s	remaining: 2.19s
517:	learn: 0.3258524	total: 2.35s	remaining: 2.19s
518:	learn: 0.3255794	total: 2.36s	remaining: 2.19s
519:	learn: 0.3251552	total: 2.36s	remaining: 2.18s
520:	learn: 0.3247569	total: 2.37s	remaining: 2.18s
521:	learn: 0.3242420	total: 2.37s	remaining: 2.17s
522:	learn: 0.3240746	total: 2.38s	remaining: 2.17s
523:	learn: 0.3239854	total: 2.38s	remaining: 2.16s
524:	learn: 0.3239415	total: 2.38s	remaining: 2.16s
525:	learn: 0.3236346	total: 2.39s	remaining: 2.15s
526:	learn: 0.3233439	total: 2.39s	remaining: 2.15s
527:	learn: 0.3232261	total: 2.4s	remaining: 2.14s
528:	learn: 0.3226489	total: 2.4s	remaining: 2.14s
529:	learn: 0.3221208	total: 2.41s	remaining: 2.13s
530:	learn: 0.3217815	total: 2.41s	remaining: 2.13s
531:	learn: 0.3216486	total: 2.42s	remaining: 2.13s
532:	learn: 0.3213358	total: 2.42s	remaining: 2.12s
533:	learn: 0.3211304	total: 2.43s	remaining: 2.12s
534:	learn: 0.3210996	total: 2.43s	remaining: 2.11s
535:	learn: 0.3207515	total: 2.44s	remaining: 2.11s
536:	learn: 0.3204738	total: 2.44s	remaining: 2.1s
537:	learn: 0.3198882	total: 2.44s	remaining: 2.1s
538:	learn: 0.3195154	total: 2.45s	remaining: 2.09s
539:	learn: 0.3192952	total: 2.45s	remaining: 2.09s
540:	learn: 0.3190280	total: 2.46s	remaining: 2.08s
541:	learn: 0.3187129	total: 2.46s	remaining: 2.08s
542:	learn: 0.3181960	total: 2.46s	remaining: 2.07s
543:	learn: 0.3176606	total: 2.47s	remaining: 2.07s
544:	learn: 0.3170949	total: 2.47s	remaining: 2.06s
545:	learn: 0.3167459	total: 2.48s	remaining: 2.06s
546:	learn: 0.3163222	total: 2.48s	remaining: 2.06s
547:	learn: 0.3161420	total: 2.49s	remaining: 2.05s
548:	learn: 0.3157992	total: 2.49s	remaining: 2.05s
549:	learn: 0.3152920	total: 2.5s	remaining: 2.04s
550:	learn: 0.3148228	total: 2.5s	remaining: 2.04s
551:	learn: 0.3145167	total: 2.5s	remaining: 2.03s
552:	learn: 0.3139810	total: 2.51s	remaining: 2.03s
553:	learn: 0.3133923	total: 2.51s	remaining: 2.02s
554:	learn: 0.3129725	total: 2.52s	remaining: 2.02s
555:	learn: 0.3125777	total: 2.52s	remaining: 2.01s
556:	learn: 0.3123591	total: 2.53s	remaining: 2.01s
557:	learn: 0.3118976	total: 2.53s	remaining: 2s
558:	learn: 0.3117266	total: 2.54s	remaining: 2s
559:	learn: 0.3114944	total: 2.54s	remaining: 2s
560:	learn: 0.3112216	total: 2.54s	remaining: 1.99s
561:	learn: 0.3108621	total: 2.55s	remaining: 1.99s
562:	learn: 0.3104885	total: 2.55s	remaining: 1.98s
563:	learn: 0.3103672	total: 2.56s	remaining: 1.98s
564:	learn: 0.3099546	total: 2.56s	remaining: 1.97s
565:	learn: 0.3094861	total: 2.57s	remaining: 1.97s
566:	learn: 0.3091734	total: 2.57s	remaining: 1.96s
567:	learn: 0.3089148	total: 2.58s	remaining: 1.96s
568:	learn: 0.3086673	total: 2.58s	remaining: 1.95s
569:	learn: 0.3082695	total: 2.58s	remaining: 1.95s
570:	learn: 0.3078891	total: 2.59s	remaining: 1.94s
571:	learn: 0.3074927	total: 2.59s	remaining: 1.94s
572:	learn: 0.3069862	total: 2.6s	remaining: 1.94s
573:	learn: 0.3064039	total: 2.6s	remaining: 1.93s
574:	learn: 0.3061550	total: 2.6s	remaining: 1.93s
575:	learn: 0.3056908	total: 2.61s	remaining: 1.92s
576:	learn: 0.3052593	total: 2.61s	remaining: 1.92s
577:	learn: 0.3049534	total: 2.62s	remaining: 1.91s
578:	learn: 0.3044688	total: 2.62s	remaining: 1.91s
579:	learn: 0.3041772	total: 2.63s	remaining: 1.9s
580:	learn: 0.3037533	total: 2.63s	remaining: 1.9s
581:	learn: 0.3035700	total: 2.64s	remaining: 1.89s
582:	learn: 0.3031629	total: 2.64s	remaining: 1.89s
583:	learn: 0.3028838	total: 2.64s	remaining: 1.88s
584:	learn: 0.3023030	total: 2.65s	remaining: 1.88s
585:	learn: 0.3019222	total: 2.65s	remaining: 1.87s
586:	learn: 0.3015915	total: 2.66s	remaining: 1.87s
587:	learn: 0.3015704	total: 2.66s	remaining: 1.86s
588:	learn: 0.3009768	total: 2.67s	remaining: 1.86s
589:	learn: 0.3008180	total: 2.67s	remaining: 1.85s
590:	learn: 0.3005118	total: 2.67s	remaining: 1.85s
591:	learn: 0.3002814	total: 2.68s	remaining: 1.85s
592:	learn: 0.3000431	total: 2.68s	remaining: 1.84s
593:	learn: 0.2997323	total: 2.69s	remaining: 1.84s
594:	learn: 0.2994389	total: 2.69s	remaining: 1.83s
595:	learn: 0.2993389	total: 2.7s	remaining: 1.83s
596:	learn: 0.2988775	total: 2.7s	remaining: 1.82s
597:	learn: 0.2987598	total: 2.7s	remaining: 1.82s
598:	learn: 0.2984033	total: 2.71s	remaining: 1.81s
599:	learn: 0.2978245	total: 2.71s	remaining: 1.81s
600:	learn: 0.2974830	total: 2.72s	remaining: 1.8s
601:	learn: 0.2969405	total: 2.72s	remaining: 1.8s
602:	learn: 0.2966278	total: 2.73s	remaining: 1.79s
603:	learn: 0.2962533	total: 2.73s	remaining: 1.79s
604:	learn: 0.2959855	total: 2.73s	remaining: 1.78s
605:	learn: 0.2956097	total: 2.74s	remaining: 1.78s
606:	learn: 0.2953299	total: 2.75s	remaining: 1.78s
607:	learn: 0.2951260	total: 2.75s	remaining: 1.77s
608:	learn: 0.2947087	total: 2.75s	remaining: 1.77s
609:	learn: 0.2944823	total: 2.76s	remaining: 1.76s
610:	learn: 0.2943654	total: 2.76s	remaining: 1.76s
611:	learn: 0.2941610	total: 2.77s	remaining: 1.75s
612:	learn: 0.2938108	total: 2.77s	remaining: 1.75s
613:	learn: 0.2934358	total: 2.78s	remaining: 1.75s
614:	learn: 0.2931197	total: 2.78s	remaining: 1.74s
615:	learn: 0.2930426	total: 2.79s	remaining: 1.74s
616:	learn: 0.2928295	total: 2.79s	remaining: 1.73s
617:	learn: 0.2924921	total: 2.79s	remaining: 1.73s
618:	learn: 0.2921751	total: 2.8s	remaining: 1.72s
619:	learn: 0.2918991	total: 2.8s	remaining: 1.72s
620:	learn: 0.2913899	total: 2.81s	remaining: 1.71s
621:	learn: 0.2908501	total: 2.81s	remaining: 1.71s
622:	learn: 0.2905626	total: 2.82s	remaining: 1.7s
623:	learn: 0.2902904	total: 2.82s	remaining: 1.7s
624:	learn: 0.2900379	total: 2.83s	remaining: 1.7s
625:	learn: 0.2896903	total: 2.83s	remaining: 1.69s
626:	learn: 0.2892638	total: 2.83s	remaining: 1.69s
627:	learn: 0.2889578	total: 2.84s	remaining: 1.68s
628:	learn: 0.2887475	total: 2.84s	remaining: 1.68s
629:	learn: 0.2883429	total: 2.85s	remaining: 1.67s
630:	learn: 0.2881113	total: 2.85s	remaining: 1.67s
631:	learn: 0.2876183	total: 2.85s	remaining: 1.66s
632:	learn: 0.2875565	total: 2.86s	remaining: 1.66s
633:	learn: 0.2871677	total: 2.86s	remaining: 1.65s
634:	learn: 0.2868290	total: 2.87s	remaining: 1.65s
635:	learn: 0.2866803	total: 2.87s	remaining: 1.64s
636:	learn: 0.2864461	total: 2.88s	remaining: 1.64s
637:	learn: 0.2861437	total: 2.88s	remaining: 1.64s
638:	learn: 0.2858220	total: 2.89s	remaining: 1.63s
639:	learn: 0.2856065	total: 2.89s	remaining: 1.63s
640:	learn: 0.2853185	total: 2.89s	remaining: 1.62s
641:	learn: 0.2849693	total: 2.9s	remaining: 1.62s
642:	learn: 0.2844873	total: 2.9s	remaining: 1.61s
643:	learn: 0.2841484	total: 2.91s	remaining: 1.61s
644:	learn: 0.2838365	total: 2.91s	remaining: 1.6s
645:	learn: 0.2835952	total: 2.92s	remaining: 1.6s
646:	learn: 0.2831927	total: 2.92s	remaining: 1.59s
647:	learn: 0.2827722	total: 2.92s	remaining: 1.59s
648:	learn: 0.2824569	total: 2.93s	remaining: 1.58s
649:	learn: 0.2821450	total: 2.93s	remaining: 1.58s
650:	learn: 0.2820499	total: 2.94s	remaining: 1.57s
651:	learn: 0.2818476	total: 2.94s	remaining: 1.57s
652:	learn: 0.2815995	total: 2.95s	remaining: 1.57s
653:	learn: 0.2812118	total: 2.95s	remaining: 1.56s
654:	learn: 0.2809202	total: 2.96s	remaining: 1.56s
655:	learn: 0.2806815	total: 2.96s	remaining: 1.55s
656:	learn: 0.2802227	total: 2.96s	remaining: 1.55s
657:	learn: 0.2800624	total: 2.97s	remaining: 1.54s
658:	learn: 0.2795849	total: 2.97s	remaining: 1.54s
659:	learn: 0.2793078	total: 2.98s	remaining: 1.53s
660:	learn: 0.2791128	total: 2.98s	remaining: 1.53s
661:	learn: 0.2788399	total: 2.99s	remaining: 1.52s
662:	learn: 0.2788037	total: 2.99s	remaining: 1.52s
663:	learn: 0.2786668	total: 3s	remaining: 1.51s
664:	learn: 0.2784946	total: 3s	remaining: 1.51s
665:	learn: 0.2780830	total: 3s	remaining: 1.51s
666:	learn: 0.2778176	total: 3.01s	remaining: 1.5s
667:	learn: 0.2775362	total: 3.01s	remaining: 1.5s
668:	learn: 0.2771948	total: 3.02s	remaining: 1.49s
669:	learn: 0.2770151	total: 3.02s	remaining: 1.49s
670:	learn: 0.2766705	total: 3.03s	remaining: 1.48s
671:	learn: 0.2761992	total: 3.03s	remaining: 1.48s
672:	learn: 0.2761667	total: 3.04s	remaining: 1.48s
673:	learn: 0.2758894	total: 3.04s	remaining: 1.47s
674:	learn: 0.2754654	total: 3.04s	remaining: 1.47s
675:	learn: 0.2751694	total: 3.05s	remaining: 1.46s
676:	learn: 0.2747818	total: 3.05s	remaining: 1.46s
677:	learn: 0.2743284	total: 3.06s	remaining: 1.45s
678:	learn: 0.2738872	total: 3.06s	remaining: 1.45s
679:	learn: 0.2735984	total: 3.07s	remaining: 1.44s
680:	learn: 0.2732869	total: 3.07s	remaining: 1.44s
681:	learn: 0.2730972	total: 3.07s	remaining: 1.43s
682:	learn: 0.2728736	total: 3.08s	remaining: 1.43s
683:	learn: 0.2726755	total: 3.08s	remaining: 1.42s
684:	learn: 0.2723927	total: 3.09s	remaining: 1.42s
685:	learn: 0.2719884	total: 3.09s	remaining: 1.42s
686:	learn: 0.2715762	total: 3.1s	remaining: 1.41s
687:	learn: 0.2713579	total: 3.1s	remaining: 1.41s
688:	learn: 0.2711750	total: 3.1s	remaining: 1.4s
689:	learn: 0.2709947	total: 3.11s	remaining: 1.4s
690:	learn: 0.2707097	total: 3.11s	remaining: 1.39s
691:	learn: 0.2704260	total: 3.12s	remaining: 1.39s
692:	learn: 0.2701645	total: 3.12s	remaining: 1.38s
693:	learn: 0.2698393	total: 3.13s	remaining: 1.38s
694:	learn: 0.2695105	total: 3.13s	remaining: 1.37s
695:	learn: 0.2691283	total: 3.14s	remaining: 1.37s
696:	learn: 0.2688368	total: 3.14s	remaining: 1.36s
697:	learn: 0.2683466	total: 3.15s	remaining: 1.36s
698:	learn: 0.2680603	total: 3.15s	remaining: 1.36s
699:	learn: 0.2677699	total: 3.15s	remaining: 1.35s
700:	learn: 0.2674562	total: 3.16s	remaining: 1.35s
701:	learn: 0.2671464	total: 3.16s	remaining: 1.34s
702:	learn: 0.2668508	total: 3.17s	remaining: 1.34s
703:	learn: 0.2666627	total: 3.17s	remaining: 1.33s
704:	learn: 0.2663736	total: 3.18s	remaining: 1.33s
705:	learn: 0.2660888	total: 3.18s	remaining: 1.32s
706:	learn: 0.2658267	total: 3.18s	remaining: 1.32s
707:	learn: 0.2656145	total: 3.19s	remaining: 1.31s
708:	learn: 0.2654194	total: 3.19s	remaining: 1.31s
709:	learn: 0.2651461	total: 3.2s	remaining: 1.3s
710:	learn: 0.2647736	total: 3.2s	remaining: 1.3s
711:	learn: 0.2644030	total: 3.21s	remaining: 1.3s
712:	learn: 0.2641493	total: 3.21s	remaining: 1.29s
713:	learn: 0.2636034	total: 3.21s	remaining: 1.29s
714:	learn: 0.2633467	total: 3.22s	remaining: 1.28s
715:	learn: 0.2630884	total: 3.22s	remaining: 1.28s
716:	learn: 0.2627231	total: 3.23s	remaining: 1.27s
717:	learn: 0.2623737	total: 3.23s	remaining: 1.27s
718:	learn: 0.2620890	total: 3.23s	remaining: 1.26s
719:	learn: 0.2619557	total: 3.24s	remaining: 1.26s
720:	learn: 0.2615330	total: 3.25s	remaining: 1.25s
721:	learn: 0.2613678	total: 3.25s	remaining: 1.25s
722:	learn: 0.2609768	total: 3.25s	remaining: 1.25s
723:	learn: 0.2607843	total: 3.26s	remaining: 1.24s
724:	learn: 0.2604017	total: 3.26s	remaining: 1.24s
725:	learn: 0.2601656	total: 3.27s	remaining: 1.23s
726:	learn: 0.2598170	total: 3.27s	remaining: 1.23s
727:	learn: 0.2596272	total: 3.27s	remaining: 1.22s
728:	learn: 0.2594368	total: 3.28s	remaining: 1.22s
729:	learn: 0.2591430	total: 3.28s	remaining: 1.21s
730:	learn: 0.2588290	total: 3.29s	remaining: 1.21s
731:	learn: 0.2585789	total: 3.29s	remaining: 1.21s
732:	learn: 0.2584455	total: 3.3s	remaining: 1.2s
733:	learn: 0.2584104	total: 3.3s	remaining: 1.2s
734:	learn: 0.2582152	total: 3.3s	remaining: 1.19s
735:	learn: 0.2581397	total: 3.31s	remaining: 1.19s
736:	learn: 0.2578583	total: 3.31s	remaining: 1.18s
737:	learn: 0.2576197	total: 3.32s	remaining: 1.18s
738:	learn: 0.2573741	total: 3.32s	remaining: 1.17s
739:	learn: 0.2571082	total: 3.33s	remaining: 1.17s
740:	learn: 0.2568704	total: 3.33s	remaining: 1.16s
741:	learn: 0.2566573	total: 3.33s	remaining: 1.16s
742:	learn: 0.2564107	total: 3.34s	remaining: 1.16s
743:	learn: 0.2561631	total: 3.35s	remaining: 1.15s
744:	learn: 0.2561312	total: 3.35s	remaining: 1.15s
745:	learn: 0.2557099	total: 3.36s	remaining: 1.14s
746:	learn: 0.2554609	total: 3.36s	remaining: 1.14s
747:	learn: 0.2552453	total: 3.36s	remaining: 1.13s
748:	learn: 0.2549054	total: 3.37s	remaining: 1.13s
749:	learn: 0.2546371	total: 3.37s	remaining: 1.12s
750:	learn: 0.2543651	total: 3.38s	remaining: 1.12s
751:	learn: 0.2537761	total: 3.38s	remaining: 1.11s
752:	learn: 0.2535814	total: 3.39s	remaining: 1.11s
753:	learn: 0.2531956	total: 3.39s	remaining: 1.11s
754:	learn: 0.2531389	total: 3.39s	remaining: 1.1s
755:	learn: 0.2529298	total: 3.4s	remaining: 1.1s
756:	learn: 0.2527376	total: 3.4s	remaining: 1.09s
757:	learn: 0.2524734	total: 3.41s	remaining: 1.09s
758:	learn: 0.2521595	total: 3.41s	remaining: 1.08s
759:	learn: 0.2519621	total: 3.42s	remaining: 1.08s
760:	learn: 0.2516846	total: 3.42s	remaining: 1.07s
761:	learn: 0.2512891	total: 3.42s	remaining: 1.07s
762:	learn: 0.2510728	total: 3.43s	remaining: 1.06s
763:	learn: 0.2507176	total: 3.43s	remaining: 1.06s
764:	learn: 0.2503362	total: 3.44s	remaining: 1.05s
765:	learn: 0.2502037	total: 3.44s	remaining: 1.05s
766:	learn: 0.2498936	total: 3.45s	remaining: 1.05s
767:	learn: 0.2496633	total: 3.45s	remaining: 1.04s
768:	learn: 0.2493097	total: 3.46s	remaining: 1.04s
769:	learn: 0.2489150	total: 3.46s	remaining: 1.03s
770:	learn: 0.2486851	total: 3.46s	remaining: 1.03s
771:	learn: 0.2483194	total: 3.47s	remaining: 1.02s
772:	learn: 0.2479421	total: 3.47s	remaining: 1.02s
773:	learn: 0.2478561	total: 3.48s	remaining: 1.01s
774:	learn: 0.2475209	total: 3.48s	remaining: 1.01s
775:	learn: 0.2471628	total: 3.48s	remaining: 1.01s
776:	learn: 0.2469314	total: 3.49s	remaining: 1s
777:	learn: 0.2466425	total: 3.49s	remaining: 997ms
778:	learn: 0.2465065	total: 3.5s	remaining: 993ms
779:	learn: 0.2461220	total: 3.5s	remaining: 988ms
780:	learn: 0.2458115	total: 3.51s	remaining: 983ms
781:	learn: 0.2456707	total: 3.51s	remaining: 979ms
782:	learn: 0.2454929	total: 3.52s	remaining: 974ms
783:	learn: 0.2451706	total: 3.52s	remaining: 970ms
784:	learn: 0.2448523	total: 3.52s	remaining: 965ms
785:	learn: 0.2445772	total: 3.53s	remaining: 961ms
786:	learn: 0.2443875	total: 3.53s	remaining: 956ms
787:	learn: 0.2441339	total: 3.54s	remaining: 952ms
788:	learn: 0.2438773	total: 3.54s	remaining: 948ms
789:	learn: 0.2438140	total: 3.55s	remaining: 943ms
790:	learn: 0.2434810	total: 3.55s	remaining: 939ms
791:	learn: 0.2429906	total: 3.56s	remaining: 934ms
792:	learn: 0.2426931	total: 3.56s	remaining: 930ms
793:	learn: 0.2424043	total: 3.56s	remaining: 925ms
794:	learn: 0.2421210	total: 3.57s	remaining: 920ms
795:	learn: 0.2419921	total: 3.57s	remaining: 916ms
796:	learn: 0.2418543	total: 3.58s	remaining: 911ms
797:	learn: 0.2418231	total: 3.58s	remaining: 907ms
798:	learn: 0.2415865	total: 3.59s	remaining: 902ms
799:	learn: 0.2412327	total: 3.59s	remaining: 898ms
800:	learn: 0.2410130	total: 3.6s	remaining: 893ms
801:	learn: 0.2406898	total: 3.6s	remaining: 889ms
802:	learn: 0.2403997	total: 3.6s	remaining: 884ms
803:	learn: 0.2403092	total: 3.61s	remaining: 880ms
804:	learn: 0.2399602	total: 3.61s	remaining: 875ms
805:	learn: 0.2397046	total: 3.62s	remaining: 870ms
806:	learn: 0.2395811	total: 3.62s	remaining: 866ms
807:	learn: 0.2393554	total: 3.62s	remaining: 861ms
808:	learn: 0.2389797	total: 3.63s	remaining: 857ms
809:	learn: 0.2386822	total: 3.63s	remaining: 852ms
810:	learn: 0.2385272	total: 3.64s	remaining: 848ms
811:	learn: 0.2383253	total: 3.64s	remaining: 843ms
812:	learn: 0.2382769	total: 3.65s	remaining: 839ms
813:	learn: 0.2380034	total: 3.65s	remaining: 834ms
814:	learn: 0.2378123	total: 3.65s	remaining: 830ms
815:	learn: 0.2376812	total: 3.66s	remaining: 825ms
816:	learn: 0.2375900	total: 3.66s	remaining: 821ms
817:	learn: 0.2373746	total: 3.67s	remaining: 817ms
818:	learn: 0.2368950	total: 3.67s	remaining: 812ms
819:	learn: 0.2365626	total: 3.68s	remaining: 808ms
820:	learn: 0.2364160	total: 3.68s	remaining: 803ms
821:	learn: 0.2362066	total: 3.69s	remaining: 798ms
822:	learn: 0.2357433	total: 3.69s	remaining: 794ms
823:	learn: 0.2355107	total: 3.69s	remaining: 789ms
824:	learn: 0.2353031	total: 3.7s	remaining: 785ms
825:	learn: 0.2349189	total: 3.7s	remaining: 780ms
826:	learn: 0.2345722	total: 3.71s	remaining: 776ms
827:	learn: 0.2342914	total: 3.71s	remaining: 771ms
828:	learn: 0.2340122	total: 3.72s	remaining: 767ms
829:	learn: 0.2337871	total: 3.72s	remaining: 762ms
830:	learn: 0.2337141	total: 3.73s	remaining: 758ms
831:	learn: 0.2334017	total: 3.73s	remaining: 754ms
832:	learn: 0.2331498	total: 3.74s	remaining: 749ms
833:	learn: 0.2329182	total: 3.74s	remaining: 745ms
834:	learn: 0.2326496	total: 3.75s	remaining: 740ms
835:	learn: 0.2323721	total: 3.75s	remaining: 736ms
836:	learn: 0.2321663	total: 3.75s	remaining: 731ms
837:	learn: 0.2317915	total: 3.76s	remaining: 727ms
838:	learn: 0.2317042	total: 3.76s	remaining: 722ms
839:	learn: 0.2315415	total: 3.77s	remaining: 718ms
840:	learn: 0.2312144	total: 3.77s	remaining: 713ms
841:	learn: 0.2309861	total: 3.78s	remaining: 709ms
842:	learn: 0.2308992	total: 3.78s	remaining: 704ms
843:	learn: 0.2307248	total: 3.78s	remaining: 699ms
844:	learn: 0.2304123	total: 3.79s	remaining: 695ms
845:	learn: 0.2302900	total: 3.79s	remaining: 690ms
846:	learn: 0.2299298	total: 3.8s	remaining: 686ms
847:	learn: 0.2297649	total: 3.8s	remaining: 681ms
848:	learn: 0.2296314	total: 3.81s	remaining: 677ms
849:	learn: 0.2292939	total: 3.81s	remaining: 672ms
850:	learn: 0.2291602	total: 3.81s	remaining: 668ms
851:	learn: 0.2288865	total: 3.82s	remaining: 663ms
852:	learn: 0.2286853	total: 3.82s	remaining: 659ms
853:	learn: 0.2285428	total: 3.83s	remaining: 654ms
854:	learn: 0.2284197	total: 3.83s	remaining: 650ms
855:	learn: 0.2282277	total: 3.83s	remaining: 645ms
856:	learn: 0.2278907	total: 3.84s	remaining: 641ms
857:	learn: 0.2277383	total: 3.84s	remaining: 636ms
858:	learn: 0.2275067	total: 3.85s	remaining: 632ms
859:	learn: 0.2272363	total: 3.85s	remaining: 627ms
860:	learn: 0.2269635	total: 3.86s	remaining: 623ms
861:	learn: 0.2265979	total: 3.86s	remaining: 618ms
862:	learn: 0.2264905	total: 3.87s	remaining: 614ms
863:	learn: 0.2262386	total: 3.87s	remaining: 609ms
864:	learn: 0.2260311	total: 3.87s	remaining: 605ms
865:	learn: 0.2258230	total: 3.88s	remaining: 600ms
866:	learn: 0.2255610	total: 3.88s	remaining: 596ms
867:	learn: 0.2254607	total: 3.89s	remaining: 591ms
868:	learn: 0.2250433	total: 3.89s	remaining: 587ms
869:	learn: 0.2249648	total: 3.9s	remaining: 582ms
870:	learn: 0.2247922	total: 3.9s	remaining: 578ms
871:	learn: 0.2244233	total: 3.9s	remaining: 573ms
872:	learn: 0.2242406	total: 3.91s	remaining: 569ms
873:	learn: 0.2240229	total: 3.91s	remaining: 564ms
874:	learn: 0.2238258	total: 3.92s	remaining: 560ms
875:	learn: 0.2237571	total: 3.92s	remaining: 555ms
876:	learn: 0.2235548	total: 3.93s	remaining: 551ms
877:	learn: 0.2234135	total: 3.93s	remaining: 546ms
878:	learn: 0.2230922	total: 3.94s	remaining: 542ms
879:	learn: 0.2226509	total: 3.94s	remaining: 537ms
880:	learn: 0.2224404	total: 3.94s	remaining: 533ms
881:	learn: 0.2223705	total: 3.95s	remaining: 528ms
882:	learn: 0.2223260	total: 3.95s	remaining: 524ms
883:	learn: 0.2220318	total: 3.96s	remaining: 519ms
884:	learn: 0.2218552	total: 3.96s	remaining: 515ms
885:	learn: 0.2217687	total: 3.97s	remaining: 510ms
886:	learn: 0.2214940	total: 3.97s	remaining: 506ms
887:	learn: 0.2213258	total: 3.98s	remaining: 501ms
888:	learn: 0.2212884	total: 3.98s	remaining: 497ms
889:	learn: 0.2210636	total: 3.98s	remaining: 493ms
890:	learn: 0.2208620	total: 3.99s	remaining: 488ms
891:	learn: 0.2207268	total: 4s	remaining: 484ms
892:	learn: 0.2205061	total: 4s	remaining: 479ms
893:	learn: 0.2203278	total: 4s	remaining: 475ms
894:	learn: 0.2200449	total: 4.01s	remaining: 470ms
895:	learn: 0.2197147	total: 4.01s	remaining: 466ms
896:	learn: 0.2195113	total: 4.02s	remaining: 461ms
897:	learn: 0.2192671	total: 4.02s	remaining: 457ms
898:	learn: 0.2192258	total: 4.03s	remaining: 452ms
899:	learn: 0.2189515	total: 4.03s	remaining: 448ms
900:	learn: 0.2186505	total: 4.03s	remaining: 443ms
901:	learn: 0.2184193	total: 4.04s	remaining: 439ms
902:	learn: 0.2183672	total: 4.04s	remaining: 434ms
903:	learn: 0.2181273	total: 4.05s	remaining: 430ms
904:	learn: 0.2178674	total: 4.05s	remaining: 425ms
905:	learn: 0.2176005	total: 4.05s	remaining: 421ms
906:	learn: 0.2174875	total: 4.06s	remaining: 416ms
907:	learn: 0.2171777	total: 4.06s	remaining: 412ms
908:	learn: 0.2169544	total: 4.07s	remaining: 407ms
909:	learn: 0.2167341	total: 4.07s	remaining: 403ms
910:	learn: 0.2166133	total: 4.08s	remaining: 398ms
911:	learn: 0.2164535	total: 4.08s	remaining: 394ms
912:	learn: 0.2164128	total: 4.08s	remaining: 389ms
913:	learn: 0.2162686	total: 4.09s	remaining: 385ms
914:	learn: 0.2160413	total: 4.09s	remaining: 380ms
915:	learn: 0.2158222	total: 4.1s	remaining: 376ms
916:	learn: 0.2157209	total: 4.1s	remaining: 371ms
917:	learn: 0.2155155	total: 4.11s	remaining: 367ms
918:	learn: 0.2154142	total: 4.11s	remaining: 362ms
919:	learn: 0.2151079	total: 4.12s	remaining: 358ms
920:	learn: 0.2149705	total: 4.12s	remaining: 353ms
921:	learn: 0.2149322	total: 4.12s	remaining: 349ms
922:	learn: 0.2148807	total: 4.13s	remaining: 344ms
923:	learn: 0.2146427	total: 4.13s	remaining: 340ms
924:	learn: 0.2145465	total: 4.14s	remaining: 335ms
925:	learn: 0.2143896	total: 4.14s	remaining: 331ms
926:	learn: 0.2141042	total: 4.15s	remaining: 327ms
927:	learn: 0.2140653	total: 4.15s	remaining: 322ms
928:	learn: 0.2138073	total: 4.16s	remaining: 318ms
929:	learn: 0.2135790	total: 4.16s	remaining: 313ms
930:	learn: 0.2135487	total: 4.16s	remaining: 309ms
931:	learn: 0.2134147	total: 4.17s	remaining: 304ms
932:	learn: 0.2132543	total: 4.17s	remaining: 300ms
933:	learn: 0.2131319	total: 4.18s	remaining: 295ms
934:	learn: 0.2128832	total: 4.18s	remaining: 291ms
935:	learn: 0.2125868	total: 4.18s	remaining: 286ms
936:	learn: 0.2124035	total: 4.19s	remaining: 282ms
937:	learn: 0.2121759	total: 4.19s	remaining: 277ms
938:	learn: 0.2120445	total: 4.2s	remaining: 273ms
939:	learn: 0.2118077	total: 4.2s	remaining: 268ms
940:	learn: 0.2114668	total: 4.21s	remaining: 264ms
941:	learn: 0.2111545	total: 4.21s	remaining: 259ms
942:	learn: 0.2111246	total: 4.21s	remaining: 255ms
943:	learn: 0.2109490	total: 4.22s	remaining: 250ms
944:	learn: 0.2107927	total: 4.22s	remaining: 246ms
945:	learn: 0.2105585	total: 4.23s	remaining: 241ms
946:	learn: 0.2105239	total: 4.23s	remaining: 237ms
947:	learn: 0.2104112	total: 4.24s	remaining: 232ms
948:	learn: 0.2101941	total: 4.24s	remaining: 228ms
949:	learn: 0.2101563	total: 4.24s	remaining: 223ms
950:	learn: 0.2101322	total: 4.25s	remaining: 219ms
951:	learn: 0.2098230	total: 4.25s	remaining: 214ms
952:	learn: 0.2097301	total: 4.26s	remaining: 210ms
953:	learn: 0.2095164	total: 4.26s	remaining: 206ms
954:	learn: 0.2093993	total: 4.27s	remaining: 201ms
955:	learn: 0.2091224	total: 4.27s	remaining: 197ms
956:	learn: 0.2089994	total: 4.28s	remaining: 192ms
957:	learn: 0.2088083	total: 4.28s	remaining: 188ms
958:	learn: 0.2087013	total: 4.29s	remaining: 183ms
959:	learn: 0.2084003	total: 4.29s	remaining: 179ms
960:	learn: 0.2082851	total: 4.29s	remaining: 174ms
961:	learn: 0.2080820	total: 4.3s	remaining: 170ms
962:	learn: 0.2078811	total: 4.3s	remaining: 165ms
963:	learn: 0.2076056	total: 4.31s	remaining: 161ms
964:	learn: 0.2074108	total: 4.31s	remaining: 156ms
965:	learn: 0.2070406	total: 4.32s	remaining: 152ms
966:	learn: 0.2067699	total: 4.32s	remaining: 148ms
967:	learn: 0.2063896	total: 4.33s	remaining: 143ms
968:	learn: 0.2060907	total: 4.33s	remaining: 139ms
969:	learn: 0.2059021	total: 4.33s	remaining: 134ms
970:	learn: 0.2056345	total: 4.34s	remaining: 130ms
971:	learn: 0.2053408	total: 4.34s	remaining: 125ms
972:	learn: 0.2050676	total: 4.35s	remaining: 121ms
973:	learn: 0.2048304	total: 4.35s	remaining: 116ms
974:	learn: 0.2046749	total: 4.36s	remaining: 112ms
975:	learn: 0.2044718	total: 4.36s	remaining: 107ms
976:	learn: 0.2041876	total: 4.37s	remaining: 103ms
977:	learn: 0.2039090	total: 4.37s	remaining: 98.3ms
978:	learn: 0.2037805	total: 4.38s	remaining: 93.9ms
979:	learn: 0.2034705	total: 4.38s	remaining: 89.4ms
980:	learn: 0.2032025	total: 4.38s	remaining: 84.9ms
981:	learn: 0.2030209	total: 4.39s	remaining: 80.4ms
982:	learn: 0.2027762	total: 4.39s	remaining: 76ms
983:	learn: 0.2026031	total: 4.4s	remaining: 71.5ms
984:	learn: 0.2023929	total: 4.4s	remaining: 67ms
985:	learn: 0.2021011	total: 4.41s	remaining: 62.6ms
986:	learn: 0.2018960	total: 4.41s	remaining: 58.1ms
987:	learn: 0.2017884	total: 4.41s	remaining: 53.6ms
988:	learn: 0.2014352	total: 4.42s	remaining: 49.1ms
989:	learn: 0.2012731	total: 4.42s	remaining: 44.7ms
990:	learn: 0.2010667	total: 4.43s	remaining: 40.2ms
991:	learn: 0.2008429	total: 4.43s	remaining: 35.7ms
992:	learn: 0.2006669	total: 4.43s	remaining: 31.3ms
993:	learn: 0.2005137	total: 4.44s	remaining: 26.8ms
994:	learn: 0.2001325	total: 4.44s	remaining: 22.3ms
995:	learn: 0.1999855	total: 4.45s	remaining: 17.9ms
996:	learn: 0.1996729	total: 4.45s	remaining: 13.4ms
997:	learn: 0.1994904	total: 4.46s	remaining: 8.93ms
998:	learn: 0.1993044	total: 4.46s	remaining: 4.47ms
999:	learn: 0.1989547	total: 4.47s	remaining: 0us


df_compare


# Train the linear regression model
lr_model_base = LinearRegression()
lr_model_base.fit(X_train_scaled, y_train)

# Make predictions on the scaled test set
lr_pred_base = lr_model_base.predict(X_test_scaled)


prediction_df = pd.DataFrame()
prediction_df['date'] = df[df.date.dt.year>=2020]['date']
prediction_df['y_test'] = y_test
prediction_df['lr_pred_base'] = lr_pred_base

prediction_df.head()


lr_score_base = evaluate_regression_model(y_test, lr_pred_base)

Mean Squared Error (MSE): 0.776
Root Mean Squared Error (RMSE): 0.881
Mean Absolute Error (MAE): 0.614
R-squared (R2): 0.981


lr_score_base

{'MSE': 0.7755799343709945,
 'RMSE': 0.880670162076015,
 'MAE': 0.6139047642970673,
 'R2': 0.9810417591179742}


plot_regression_accuracy(y_test, lr_pred_base)


plot_predictions(df,lr_pred_base)


lr_base_feature_importance = plot_feature_importance(lr_model_base,X_train,20)


lr_base_feature_importance[:15]


keep_cols20 = lr_base_feature_importance[:20]['Feature'].tolist()

X_train20 = X_train[keep_cols20]
X_test20 = X_test[keep_cols20]

scaler = StandardScaler()
X_train_scaled20 = scaler.fit_transform(X_train20)
X_test_scaled20 = scaler.transform(X_test20)


# Train the linear regression model
lr_model20 = LinearRegression()
lr_model20.fit(X_train_scaled20, y_train)

# Make predictions on the scaled test set
lr_pred20 = lr_model20.predict(X_test_scaled20)
lr_score20 = evaluate_regression_model(y_test, lr_pred20)

Mean Squared Error (MSE): 0.768
Root Mean Squared Error (RMSE): 0.877
Mean Absolute Error (MAE): 0.613
R-squared (R2): 0.981


prediction_df['lr_pred20'] = lr_pred20

prediction_df.head()


lr_score20

{'MSE': 0.7682885573217368,
 'RMSE': 0.8765207112908039,
 'MAE': 0.6133462584657988,
 'R2': 0.9812199892092073}


plot_feature_importance(lr_model20,X_train20,20)


keep_cols15 = lr_base_feature_importance[:15]['Feature'].tolist()

X_train15 = X_train[keep_cols15]
X_test15 = X_test[keep_cols15]

scaler = StandardScaler()
X_train_scaled15 = scaler.fit_transform(X_train15)
X_test_scaled15 = scaler.transform(X_test15)


# Train the linear regression model
lr_model15 = LinearRegression()
lr_model15.fit(X_train_scaled15, y_train)

# Make predictions on the scaled test set
lr_pred15 = lr_model15.predict(X_test_scaled15)
lr_score15 = evaluate_regression_model(y_test, lr_pred15)

Mean Squared Error (MSE): 0.797
Root Mean Squared Error (RMSE): 0.893
Mean Absolute Error (MAE): 0.626
R-squared (R2): 0.981


prediction_df['lr_pred15'] = lr_pred15

prediction_df.head()


lr_score15

{'MSE': 0.7968925999367115,
 'RMSE': 0.8926884114497686,
 'MAE': 0.6261544545364333,
 'R2': 0.9805207932836008}


plot_feature_importance(lr_model15,X_train15,15)


keep_cols10 = lr_base_feature_importance[:10]['Feature'].tolist()

X_train10 = X_train[keep_cols10]
X_test10 = X_test[keep_cols10]

scaler = StandardScaler()
X_train_scaled10 = scaler.fit_transform(X_train10)
X_test_scaled10 = scaler.transform(X_test10)


# Train the linear regression model
lr_model10 = LinearRegression()
lr_model10.fit(X_train_scaled10, y_train)

# Make predictions on the scaled test set
lr_pred10 = lr_model10.predict(X_test_scaled10)
lr_score10 = evaluate_regression_model(y_test, lr_pred10)

Mean Squared Error (MSE): 0.786
Root Mean Squared Error (RMSE): 0.887
Mean Absolute Error (MAE): 0.626
R-squared (R2): 0.981


prediction_df['lr_pred10'] = lr_pred10

prediction_df.head()


lr_score10

{'MSE': 0.7864993593280865,
 'RMSE': 0.8868479911056271,
 'MAE': 0.6260216655405225,
 'R2': 0.9807748451875646}


plot_feature_importance(lr_model10,X_train10,10)


ridge_model = Ridge()

    # Define the hyperparameter grid to search
param_grid = {'alpha': [0.001, 0.01, 0.1, 1, 10, 100]}

    # Perform GridSearchCV for hyperparameter tuning
grid_search = GridSearchCV(estimator=ridge_model, param_grid=param_grid, scoring='neg_mean_squared_error', cv=5)
grid_search.fit(X_train_scaled, y_train)

    # Get the best model
best_ridge_model = grid_search.best_estimator_

    # Make predictions on the test set
ridge_pred_base = best_ridge_model.predict(X_test_scaled)

    # Evaluate the best model
mse = mean_squared_error(y_test, ridge_pred_base)
rmse = mean_squared_error(y_test, ridge_pred_base, squared=False)
mae = mean_absolute_error(y_test, ridge_pred_base)
r2 = r2_score(y_test, ridge_pred_base)

print("Best Ridge Regression Model:")
print(f"Best alpha: {best_ridge_model.alpha}")
print(f'Root Mean Squared Error (RMSE): {np.round(rmse,3)}')
print(f"Mean Squared Error: {np.round(mse,3)}")
print(f"Mean Absolute Error: {np.round(mae,3)}")
print(f"R2 Score: {np.round(r2,3)}")

ridge_score = {
        'MSE': mse,
        'RMSE': rmse,
        'MAE': mae,
        'R2': r2
    }

Best Ridge Regression Model:
Best alpha: 0.001
Root Mean Squared Error (RMSE): 0.871
Mean Squared Error: 0.759
Mean Absolute Error: 0.606
R2 Score: 0.981


ridge_base_feature_importance = plot_feature_importance(best_ridge_model,X_train,20)


ridge_base_feature_importance[:20]


prediction_df['ridge_pred_base'] = ridge_pred_base

prediction_df.head()


keep_cols20 = ridge_base_feature_importance[:20]['Feature'].tolist()

X_train20 = X_train[keep_cols20]
X_test20 = X_test[keep_cols20]

scaler = StandardScaler()
X_train_scaled20 = scaler.fit_transform(X_train20)
X_test_scaled20 = scaler.transform(X_test20)

# Train model
ridge_model20 = Ridge(alpha=0.001)
ridge_model20.fit(X_train_scaled20, y_train)


# Make predictions on the scaled test set
ridge_pred20 = ridge_model20.predict(X_test_scaled20)
ridge_score20 = evaluate_regression_model(y_test, ridge_pred20)

Mean Squared Error (MSE): 0.73
Root Mean Squared Error (RMSE): 0.854
Mean Absolute Error (MAE): 0.586
R-squared (R2): 0.982


plot_feature_importance(ridge_model20,X_train20,20)


prediction_df['ridge_pred20'] = ridge_pred20

prediction_df.head()


lasso_model = Lasso()

param_grid = {'alpha': [0.001, 0.01, 0.1, 1, 10, 100]}

    # Perform GridSearchCV for hyperparameter tuning
grid_search = GridSearchCV(estimator=lasso_model, param_grid=param_grid, scoring='neg_mean_squared_error', cv=5)
grid_search.fit(X_train_scaled, y_train)

    # Get the best model
best_lasso_model = grid_search.best_estimator_

    # Make predictions on the test set
lasso_pred_base = best_lasso_model.predict(X_test_scaled)

    # Evaluate the best model
mse = mean_squared_error(y_test, lasso_pred_base)
rmse = mean_squared_error(y_test, lasso_pred_base, squared=False)
mae = mean_absolute_error(y_test, lasso_pred_base)
r2 = r2_score(y_test, lasso_pred_base)

print("Best Lasso Regression Model:")
print(f"Best alpha: {best_lasso_model.alpha}")
print(f'Root Mean Squared Error (RMSE): {np.round(rmse,3)}')
print(f"Mean Squared Error: {np.round(mse,3)}")
print(f"Mean Absolute Error: {np.round(mae,3)}")
print(f"R2 Score: {np.round(r2,3)}")

lasso_score = {
        'MSE': mse,
        'RMSE': rmse,
        'MAE': mae,
        'R2': r2
    }

Best Lasso Regression Model:
Best alpha: 0.001
Root Mean Squared Error (RMSE): 0.97
Mean Squared Error: 0.94
Mean Absolute Error: 0.663
R2 Score: 0.977


lasso_base_feature_importance = plot_feature_importance(best_lasso_model,X_train,20)


lasso_base_feature_importance[:20]


prediction_df['lasso_pred_base'] = lasso_pred_base

prediction_df.head()


keep_cols20 = lasso_base_feature_importance[:20]['Feature'].tolist()

X_train20 = X_train[keep_cols20]
X_test20 = X_test[keep_cols20]

scaler = StandardScaler()
X_train_scaled20 = scaler.fit_transform(X_train20)
X_test_scaled20 = scaler.transform(X_test20)

# Train model
lasso_model20 = Lasso(alpha=0.001)
lasso_model20.fit(X_train_scaled20, y_train)


# Make predictions on the scaled test set
lasso_pred20 = lasso_model20.predict(X_test_scaled20)
lasso_score20 = evaluate_regression_model(y_test, lasso_pred20)

Mean Squared Error (MSE): 0.948
Root Mean Squared Error (RMSE): 0.974
Mean Absolute Error (MAE): 0.665
R-squared (R2): 0.977


plot_feature_importance(lasso_model20,X_train20,20)


prediction_df['lasso_pred20'] = lasso_pred20

prediction_df.head()


elastic_net_model = ElasticNet()

    # Define the hyperparameter grid to search
param_grid = {
        'alpha': [0.001, 0.01, 0.1, 1, 10, 100],
        'l1_ratio': [0.1, 0.3, 0.5, 0.7, 0.9]
    }

    # Perform GridSearchCV for hyperparameter tuning
grid_search = GridSearchCV(estimator=elastic_net_model, param_grid=param_grid, scoring='neg_mean_squared_error', cv=5)
grid_search.fit(X_train_scaled, y_train)

    # Get the best model
best_elastic_net_model = grid_search.best_estimator_

    # Make predictions on the test set
elastic_pred_base = best_elastic_net_model.predict(X_test_scaled)

    # Evaluate the best model
mse = mean_squared_error(y_test, elastic_pred_base)
rmse = mean_squared_error(y_test, elastic_pred_base, squared=False)
mae = mean_absolute_error(y_test, elastic_pred_base)
r2 = r2_score(y_test, elastic_pred_base)

print("Best Elastic Net Model:")
print(f"Best alpha: {best_elastic_net_model.alpha}")
print(f"Best l1_ratio: {best_elastic_net_model.l1_ratio}")
print(f'Root Mean Squared Error (RMSE): {np.round(rmse,3)}')
print(f"Mean Squared Error: {np.round(mse,3)}")
print(f"Mean Absolute Error: {np.round(mae,3)}")
print(f"R2 Score: {np.round(r2,3)}")

elastic_score = {
        'MSE': mse,
        'RMSE': rmse,
        'MAE': mae,
        'R2': r2
    }

Best Elastic Net Model:
Best alpha: 0.001
Best l1_ratio: 0.1
Root Mean Squared Error (RMSE): 0.953
Mean Squared Error: 0.908
Mean Absolute Error: 0.653
R2 Score: 0.978


elastic_base_feature_importance = plot_feature_importance(best_elastic_net_model,X_train,20)


elastic_base_feature_importance[:20]


prediction_df['elastic_pred_base'] = elastic_pred_base

prediction_df.head()


keep_cols20 = elastic_base_feature_importance[:20]['Feature'].tolist()

X_train20 = X_train[keep_cols20]
X_test20 = X_test[keep_cols20]

scaler = StandardScaler()
X_train_scaled20 = scaler.fit_transform(X_train20)
X_test_scaled20 = scaler.transform(X_test20)

# Train model
elastic_model20 = ElasticNet(alpha=0.001,l1_ratio =  0.9)
elastic_model20.fit(X_train_scaled20, y_train)


# Make predictions on the scaled test set
elastic_pred20 = elastic_model20.predict(X_test_scaled20)
elastic_score20 = evaluate_regression_model(y_test, elastic_pred20)

Mean Squared Error (MSE): 0.948
Root Mean Squared Error (RMSE): 0.974
Mean Absolute Error (MAE): 0.667
R-squared (R2): 0.977


plot_feature_importance(elastic_model20,X_train20,20)


prediction_df['elastic_pred20'] = elastic_pred20

prediction_df.head()


ela_df = pd.DataFrame([elastic_score.keys(),elastic_score.values()])
ela_df.columns = ela_df.iloc[0]
ela_df = ela_df[1:].reset_index(drop=True)
ela_df['Model'] = 'Elastic_Net with All Features'

ela_20_df = pd.DataFrame([elastic_score20.keys(),elastic_score20.values()])
ela_20_df.columns = ela_20_df.iloc[0]
ela_20_df = ela_20_df[1:].reset_index(drop=True)
ela_20_df['Model'] = 'Elastic_Net with Top 20 Features'

lasso_df = pd.DataFrame([lasso_score.keys(),lasso_score.values()])
lasso_df.columns = lasso_df.iloc[0]
lasso_df = lasso_df[1:].reset_index(drop=True)
lasso_df['Model'] = 'Lasso with All Features'

lasso_20_df = pd.DataFrame([lasso_score20.keys(),lasso_score20.values()])
lasso_20_df.columns = lasso_20_df.iloc[0]
lasso_20_df = lasso_20_df[1:].reset_index(drop=True)
lasso_20_df['Model'] = 'Lasso with Top 20 Features'

ridge_df = pd.DataFrame([ridge_score.keys(),ridge_score.values()])
ridge_df.columns = ridge_df.iloc[0]
ridge_df = ridge_df[1:].reset_index(drop=True)
ridge_df['Model'] = 'Ridge with All Features'

ridge_20_df = pd.DataFrame([ridge_score20.keys(),ridge_score20.values()])
ridge_20_df.columns = ridge_20_df.iloc[0]
ridge_20_df = ridge_20_df[1:].reset_index(drop=True)
ridge_20_df['Model'] = 'Ridge with Top 20 Features'

lr_base_df = pd.DataFrame([lr_score_base.keys(),lr_score_base.values()])
lr_base_df.columns = lr_base_df.iloc[0]
lr_base_df = lr_base_df[1:].reset_index(drop=True)
lr_base_df['Model'] = 'Linear Reg. with All Features'

lr_20_df = pd.DataFrame([lr_score20.keys(),lr_score20.values()])
lr_20_df.columns = lr_20_df.iloc[0]
lr_20_df = lr_20_df[1:].reset_index(drop=True)
lr_20_df['Model'] = 'Linear Reg. with Top 20 Features'

lr_15_df = pd.DataFrame([lr_score15.keys(),lr_score15.values()])
lr_15_df.columns = lr_15_df.iloc[0]
lr_15_df = lr_15_df[1:].reset_index(drop=True)
lr_15_df['Model'] = 'Linear Reg. with Top 15 Features'

lr_10_df = pd.DataFrame([lr_score10.keys(),lr_score10.values()])
lr_10_df.columns = lr_10_df.iloc[0]
lr_10_df = lr_10_df[1:].reset_index(drop=True)
lr_10_df['Model'] = 'Linear Reg. with Top 10 Features'

df_compare = pd.concat([ela_df,lasso_df,ridge_df,ela_20_df,lasso_20_df,ridge_20_df,
                        lr_base_df,lr_20_df,lr_15_df,lr_10_df]).sort_values(by=['R2'],ascending=False).reset_index(drop=True)

df_compare


prediction_df


plt.figure(figsize=(20, 10))
sns.lineplot(x = prediction_df.date, y=prediction_df.y_test,label='y_test')

sns.lineplot(x = prediction_df.date, y=prediction_df.lr_pred_base,label='lr_pred_base')
sns.lineplot(x = prediction_df.date, y=prediction_df.lr_pred20,label='lr_pred20')
sns.lineplot(x = prediction_df.date, y=prediction_df.lr_pred15,label='lr_pred15')
sns.lineplot(x = prediction_df.date, y=prediction_df.lr_pred10,label='lr_pred10')

sns.lineplot(x = prediction_df.date, y=prediction_df.ridge_pred_base,label='ridge_pred_base')
sns.lineplot(x = prediction_df.date, y=prediction_df.ridge_pred20,label='ridge_pred20')

sns.lineplot(x = prediction_df.date, y=prediction_df.lasso_pred_base,label='lasso_pred_base')
sns.lineplot(x = prediction_df.date, y=prediction_df.lasso_pred20,label='lasso_pred20')

sns.lineplot(x = prediction_df.date, y=prediction_df.elastic_pred_base,label='elastic_pred_base')
sns.lineplot(x = prediction_df.date, y=prediction_df.elastic_pred20,label='elastic_pred20')

plt.legend(prop={'size': 14, 'weight': 'bold'})
plt.title('Model Prediction Comparison', fontsize=16)
plt.ylabel('Prediction', fontsize=14)
plt.xlabel('Date', fontsize=14)

plt.show()


plt.figure(figsize=(20, 10))
sns.lineplot(x = prediction_df.date, y=prediction_df.y_test,label='y_test')

sns.lineplot(x = prediction_df.date, y=prediction_df.lr_pred_base,label='lr_pred_base')
sns.lineplot(x = prediction_df.date, y=prediction_df.lr_pred20,label='lr_pred20')
sns.lineplot(x = prediction_df.date, y=prediction_df.lr_pred15,label='lr_pred15')
sns.lineplot(x = prediction_df.date, y=prediction_df.lr_pred10,label='lr_pred10')

plt.legend(prop={'size': 14, 'weight': 'bold'})
plt.title('Model Prediction Comparison', fontsize=16)
plt.ylabel('Prediction', fontsize=14)
plt.xlabel('Date', fontsize=14)

plt.show()


plt.figure(figsize=(20, 10))
sns.lineplot(x = prediction_df.date, y=prediction_df.y_test,label='y_test')

sns.lineplot(x = prediction_df.date, y=prediction_df.ridge_pred_base,label='ridge_pred_base')
sns.lineplot(x = prediction_df.date, y=prediction_df.ridge_pred20,label='ridge_pred20')

plt.legend(prop={'size': 14, 'weight': 'bold'})
plt.title('Model Prediction Comparison', fontsize=16)
plt.ylabel('Prediction', fontsize=14)
plt.xlabel('Date', fontsize=14)

plt.show()


plt.figure(figsize=(20, 10))
sns.lineplot(x = prediction_df.date, y=prediction_df.y_test,label='y_test')

sns.lineplot(x = prediction_df.date, y=prediction_df.lasso_pred_base,label='lasso_pred_base')
sns.lineplot(x = prediction_df.date, y=prediction_df.lasso_pred20,label='lasso_pred20')

plt.legend(prop={'size': 14, 'weight': 'bold'})
plt.title('Model Prediction Comparison', fontsize=16)
plt.ylabel('Prediction', fontsize=14)
plt.xlabel('Date', fontsize=14)

plt.show()


plt.figure(figsize=(20, 10))
sns.lineplot(x = prediction_df.date, y=prediction_df.y_test,label='y_test')

sns.lineplot(x = prediction_df.date, y=prediction_df.elastic_pred_base,label='elastic_pred_base')
sns.lineplot(x = prediction_df.date, y=prediction_df.elastic_pred20,label='elastic_pred20')

plt.legend(prop={'size': 14, 'weight': 'bold'})
plt.title('Model Prediction Comparison', fontsize=16)
plt.ylabel('Prediction', fontsize=14)
plt.xlabel('Date', fontsize=14)

plt.show()


# target column is next day's close price
y_train = train_df['close_1d_next'].copy()
X_train = train_df.drop(['close_1d_next'], 1)

# target column is next day's close price
y_test  = test_df['close_1d_next'].copy()
X_test  = test_df.drop(['close_1d_next'], 1)


ridge_20_features = ridge_base_feature_importance[:20]['Feature'].tolist()
X_train = X_train[ridge_20_features]
X_test = X_test[ridge_20_features]


def train_ridge_regression(X_train,X_test,y_train,y_test):
    
    scaler = StandardScaler()
    X_train_scaled = scaler.fit_transform(X_train)
    X_test_scaled = scaler.transform(X_test)

    # Train model
    ridge_model = Ridge(alpha=0.001)
    ridge_model.fit(X_train_scaled, y_train)


    # Make predictions on the scaled test set
    ridge_pred = ridge_model.predict(X_test_scaled)
    ridge_score = evaluate_regression_model2(y_test, ridge_pred)

    return ridge_model,ridge_pred,ridge_score


ridge_model, ridge_pred, ridge_score = train_ridge_regression(X_train,X_test,y_train,y_test)


ridge_score

{'MSE': 0.7295114881722916,
 'RMSE': 0.8541144467647714,
 'MAE': 0.5856010765186319,
 'R2': 0.9821678541358965}


ridge_pred[:15]

array([54.34934416, 54.38159609, 54.23318706, 54.03650291, 54.05964943,
       54.17549586, 54.25017275, 54.54902816, 54.18781481, 54.74202716,
       55.14666847, 55.18770245, 55.53205975, 55.4774224 , 55.51006433])


plot_regression_accuracy(y_test, ridge_pred)


plot_predictions(df,ridge_pred)


plot_feature_importance(ridge_model,X_train,20)


def preprocess_data(df):
    df['ema_9'] = df['close'].ewm(9).mean().shift()
    df['sma_5'] = df['close'].rolling(5).mean().shift()
    df['sma_10'] = df['close'].rolling(10).mean().shift()
    df['sma_15'] = df['close'].rolling(15).mean().shift()
    df['sma_30'] = df['close'].rolling(30).mean().shift()
    df['rsi'] = rsi(df) #.fillna(0)
    df['mfi'] = mfi(df, 14)
    EMA_12 = pd.Series(df['close'].ewm(span=12, min_periods=12).mean())
    EMA_26 = pd.Series(df['close'].ewm(span=26, min_periods=26).mean())
    df['macd'] = pd.Series(EMA_12 - EMA_26)
    df['macd_signal'] = pd.Series(df.macd.ewm(span=9, min_periods=9).mean())
    df['close_1d_next'] = df['close'].shift(-1)

    df['close_1d_ago'] = df['close'].shift(1)
    df['close_3d_ago'] = df['close'].shift(3)
    df['close_5d_ago'] = df['close'].shift(5)
    df['close_1w_ago'] = df['close'].shift(7)
    df['close_2w_ago'] = df['close'].shift(14)
    df['close_3w_ago'] = df['close'].shift(21)
    df['close_4w_ago'] = df['close'].shift(28)

    df['adj_close_1d_ago'] = df['adj close'].shift(1)
    df['adj_close_3d_ago'] = df['adj close'].shift(3)
    df['adj_close_5d_ago'] = df['adj close'].shift(5)
    df['adj_close_1w_ago'] = df['adj close'].shift(7)
    df['adj_close_2w_ago'] = df['adj close'].shift(14)
    df['adj_close_3w_ago'] = df['adj close'].shift(21)
    df['adj_close_4w_ago'] = df['adj close'].shift(28)

    df['open_1d_ago'] = df['open'].shift(1)
    df['open_3d_ago'] = df['open'].shift(3)
    df['open_5d_ago'] = df['open'].shift(5)
    df['open_1w_ago'] = df['open'].shift(7)
    df['open_2w_ago'] = df['open'].shift(14)
    df['open_3w_ago'] = df['open'].shift(21)
    df['open_4w_ago'] = df['open'].shift(28)

    df['high_1d_ago'] = df['high'].shift(1)
    df['high_3d_ago'] = df['high'].shift(3)
    df['high_5d_ago'] = df['high'].shift(5)
    df['high_1w_ago'] = df['high'].shift(7)
    df['high_2w_ago'] = df['high'].shift(14)
    df['high_3w_ago'] = df['high'].shift(21)
    df['high_4w_ago'] = df['high'].shift(28)

    df['low_1d_ago'] = df['low'].shift(1)
    df['low_3d_ago'] = df['low'].shift(3)
    df['low_5d_ago'] = df['low'].shift(5)
    df['low_1w_ago'] = df['low'].shift(7)
    df['low_2w_ago'] = df['low'].shift(14)
    df['low_3w_ago'] = df['low'].shift(21)
    df['low_4w_ago'] = df['low'].shift(28)

    df['volume_1d_ago'] = df['volume'].shift(1)
    df['volume_3d_ago'] = df['volume'].shift(3)
    df['volume_5d_ago'] = df['volume'].shift(5)
    df['volume_1w_ago'] = df['volume'].shift(7)
    df['volume_2w_ago'] = df['volume'].shift(14)
    df['volume_3w_ago'] = df['volume'].shift(21)
    df['volume_4w_ago'] = df['volume'].shift(28)

    df['open_3d_avg'] = df['open'].rolling(window=3).mean()
    df['open_5d_avg'] = df['open'].rolling(window=5).mean()
    df['open_7d_avg'] = df['open'].rolling(window=7).mean()
    df['open_10d_avg'] = df['open'].rolling(window=10).mean()
    df['open_15d_avg'] = df['open'].rolling(window=15).mean()
    df['open_30d_avg'] = df['open'].rolling(window=30).mean()

    df['high_3d_avg'] = df['high'].rolling(window=3).mean()
    df['high_5d_avg'] = df['high'].rolling(window=5).mean()
    df['high_7d_avg'] = df['high'].rolling(window=7).mean()
    df['high_10d_avg'] = df['high'].rolling(window=10).mean()
    df['high_15d_avg'] = df['high'].rolling(window=15).mean()
    df['high_30d_avg'] = df['high'].rolling(window=30).mean()

    df['low_3d_avg'] = df['low'].rolling(window=3).mean()
    df['low_5d_avg'] = df['low'].rolling(window=5).mean()
    df['low_7d_avg'] = df['low'].rolling(window=7).mean()
    df['low_10d_avg'] = df['low'].rolling(window=10).mean()
    df['low_15d_avg'] = df['low'].rolling(window=15).mean()
    df['low_30d_avg'] = df['low'].rolling(window=30).mean()

    df['volume_3d_avg'] = df['volume'].rolling(window=3).mean()
    df['volume_5d_avg'] = df['volume'].rolling(window=5).mean()
    df['volume_7d_avg'] = df['volume'].rolling(window=7).mean()
    df['volume_10d_avg'] = df['volume'].rolling(window=10).mean()
    df['volume_15d_avg'] = df['volume'].rolling(window=15).mean()
    df['volume_30d_avg'] = df['volume'].rolling(window=30).mean()

    df['adj_close_3d_avg'] = df['adj close'].rolling(window=3).mean()
    df['adj_close_5d_avg'] = df['adj close'].rolling(window=5).mean()
    df['adj_close_7d_avg'] = df['adj close'].rolling(window=7).mean()
    df['adj_close_10d_avg'] = df['adj close'].rolling(window=10).mean()
    df['adj_close_15d_avg'] = df['adj close'].rolling(window=15).mean()
    df['adj_close_30d_avg'] = df['adj close'].rolling(window=30).mean()
    
    return df


df_all = pd.read_parquet(out_loc+"stock_1d.parquet")
df_all.columns = df_all.columns.str.lower()


### keep stocks in data with min year 2013, max year 2023
stock_min_dt = pd.DataFrame(df_all.groupby('symbol')['date'].min()).reset_index().rename(columns={'date':'min_date'})
stock_max_dt = pd.DataFrame(df_all.groupby('symbol')['date'].max()).reset_index().rename(columns={'date':'max_date'})
stock_cnt_dt = pd.DataFrame(df_all.groupby('symbol')['date'].count()).reset_index().rename(columns={'date':'days_cnt'})

stock_cnt = stock_min_dt.merge(stock_max_dt,on='symbol').merge(stock_cnt_dt,on='symbol')
stock_cnt['min_year'] = stock_cnt['min_date'].dt.year
stock_cnt['max_year'] = stock_cnt['max_date'].dt.year

keep_stocks = stock_cnt[(stock_cnt['min_year']==2013)&(stock_cnt['max_year']==2023)&(stock_cnt['days_cnt']>=2500)]['symbol'].unique().tolist()

stock_cnt.head()


df_2023 = df_all[(df_all.date.dt.year==2023) & (df_all.symbol.isin(keep_stocks))]
# volume vs stocks
volume_2023 = pd.DataFrame(df_2023.groupby(['symbol','security','gics sector'])['volume'].sum()).reset_index()
volume_2023 = volume_2023.sort_values(by='volume',ascending=False).reset_index(drop=True)
volume_2023.head()


# volume vs sectors
sector_2023 = pd.DataFrame(df_2023.groupby(['gics sector'])['volume'].sum()).reset_index()
sector_2023 = sector_2023.sort_values(by='volume',ascending=False).reset_index(drop=True)
sector_2023


# filter top 5 sectors with highest volume in 2023
sector_list = sector_2023[:5]['gics sector'].tolist()

stock_list = []

num_stocks = 5
# stocks with highest volume in each sector
for sec in sector_list:
    stock_list.append(volume_2023[volume_2023['gics sector']==sec]['symbol'][:num_stocks].tolist())
    
    
stock_list = [item for sublist in stock_list for item in sublist]

len(stock_list)

25


df_stocks = df_all[df_all['symbol'].isin(stock_list)].reset_index(drop=True)
df_stocks.head()


stock_compare = []

for stock in stock_list:
        stock_data = df_stocks[df_stocks['symbol'] == stock]
        stock_data = preprocess_data(stock_data)
        stock_data = stock_data.dropna().reset_index(drop=True)
        
                # Split the DataFrame into training and testing sets
        train_df_temp = stock_data[stock_data.date.dt.year<2020]
        test_df_temp = stock_data[stock_data.date.dt.year>=2020]

        drop_cols1 = ['date','open','high','low','close','adj close','volume','symbol','security',
         'gics sector','gics sub-industry','headquarters location','date added','cik','founded']

        train_df_temp = train_df_temp.drop(drop_cols1, 1)

        test_df_temp  = test_df_temp.drop(drop_cols1, 1)

        # target column is next day's close price
        y_train_temp = train_df_temp['close_1d_next'].copy()
        X_train_temp = train_df_temp.drop(['close_1d_next'], 1)

        # target column is next day's close price
        y_test_temp  = test_df_temp['close_1d_next'].copy()
        X_test_temp  = test_df_temp.drop(['close_1d_next'], 1)
#        print(stock, len(X_train), len(X_test), len(y_train), len(y_test))
        temp_model, temp_pred, temp_score = train_ridge_regression(X_train_temp,X_test_temp,y_train_temp,y_test_temp)
        
        score_df = pd.DataFrame([temp_score.keys(),temp_score.values()])
        score_df.columns = score_df.iloc[0]
        score_df = score_df[1:].reset_index(drop=True)
        score_df['symbol'] = stock
        
        stock_compare.append(score_df)
        
compare_df = pd.concat(stock_compare).sort_values(by='R2',ascending=False).reset_index(drop =True)


compare_df

	date	open	high	low	close	adj close	volume	symbol	security	gics sector	gics sub-industry	headquarters location	date added	cik	founded
0	2013-01-02	94.190002	94.790001	93.959999	94.779999	67.895119	3206700.0	MMM	3M	Industrials	Industrial Conglomerates	Saint Paul, Minnesota	1957-03-04	66740	1902
1	2013-01-03	94.339996	94.930000	94.129997	94.669998	67.816322	2704600.0	MMM	3M	Industrials	Industrial Conglomerates	Saint Paul, Minnesota	1957-03-04	66740	1902
2	2013-01-04	94.790001	95.480003	94.540001	95.370003	68.317757	2704900.0	MMM	3M	Industrials	Industrial Conglomerates	Saint Paul, Minnesota	1957-03-04	66740	1902
3	2013-01-07	95.019997	95.730003	94.760002	95.489998	68.403717	2745800.0	MMM	3M	Industrials	Industrial Conglomerates	Saint Paul, Minnesota	1957-03-04	66740	1902
4	2013-01-08	95.169998	95.750000	95.099998	95.500000	68.410889	2655500.0	MMM	3M	Industrials	Industrial Conglomerates	Saint Paul, Minnesota	1957-03-04	66740	1902

	date	open	high	low	close	adj close	volume	symbol	security	gics sector	gics sub-industry	headquarters location	date added	cik	founded
852843	2013-01-02	25.840000	26.690001	25.780001	26.670000	21.445908	17862400.0	MDLZ	Mondelez International	Consumer Staples	Packaged Foods & Meats	Chicago, Illinois	2012-10-02	1103982	2012
852844	2013-01-03	26.700001	26.770000	26.490000	26.639999	21.421791	9075500.0	MDLZ	Mondelez International	Consumer Staples	Packaged Foods & Meats	Chicago, Illinois	2012-10-02	1103982	2012
852845	2013-01-04	26.700001	26.830000	26.549999	26.740000	21.502203	7696000.0	MDLZ	Mondelez International	Consumer Staples	Packaged Foods & Meats	Chicago, Illinois	2012-10-02	1103982	2012
852846	2013-01-07	26.620001	26.740000	26.549999	26.660000	21.437866	7576200.0	MDLZ	Mondelez International	Consumer Staples	Packaged Foods & Meats	Chicago, Illinois	2012-10-02	1103982	2012
852847	2013-01-08	26.520000	26.920000	26.459999	26.680000	21.453959	14360800.0	MDLZ	Mondelez International	Consumer Staples	Packaged Foods & Meats	Chicago, Illinois	2012-10-02	1103982	2012

	date	close	ema_9	sma_5	sma_10	sma_15	sma_30	rsi	mfi
852843	2013-01-02	26.670000	NaN	NaN	NaN	NaN	NaN	NaN	0.000000
852844	2013-01-03	26.639999	26.670000	NaN	NaN	NaN	NaN	NaN	33.904295
852845	2013-01-04	26.740000	26.653333	NaN	NaN	NaN	NaN	NaN	48.695375
852846	2013-01-07	26.660000	26.688852	NaN	NaN	NaN	NaN	NaN	39.919745
852847	2013-01-08	26.680000	26.679078	NaN	NaN	NaN	NaN	NaN	55.233142
...	...	...	...	...	...	...	...	...	...
855571	2023-11-02	67.970001	65.583879	65.934001	65.321001	64.406001	65.993000	60.257764	89.207420
855572	2023-11-03	68.820000	66.061103	66.398001	65.697001	64.868001	65.901667	63.726091	89.458580
855573	2023-11-06	68.239998	66.612883	67.160001	66.169001	65.354001	65.848000	59.885606	83.710782
855574	2023-11-07	68.489998	66.938306	67.612000	66.594001	65.728667	65.799000	60.977252	75.937617
855575	2023-11-08	69.019997	67.248644	68.067999	66.888000	66.058667	65.729667	63.259914	75.566164

	date	open	high	low	close	adj close	volume	symbol	security	gics sector	gics sub-industry	headquarters location	date added	cik	founded	ema_9	sma_5	sma_10	sma_15	sma_30	rsi	mfi	macd	macd_signal	close_1d_next	close_1d_ago	close_3d_ago	close_5d_ago	close_7d_ago	close_14d_ago	close_21d_ago	close_28d_ago	close_3d_avg	close_5d_avg	close_7d_avg	close_10d_avg	close_15d_avg	close_30d_avg	adj close_1d_ago	adj close_3d_ago	adj close_5d_ago	adj close_7d_ago	adj close_14d_ago	adj close_21d_ago	adj close_28d_ago	adj close_3d_avg	adj close_5d_avg	adj close_7d_avg	adj close_10d_avg	adj close_15d_avg	adj close_30d_avg	open_1d_ago	open_3d_ago	open_5d_ago	open_7d_ago	open_14d_ago	open_21d_ago	open_28d_ago	open_3d_avg	open_5d_avg	open_7d_avg	open_10d_avg	open_15d_avg	open_30d_avg	high_1d_ago	high_3d_ago	high_5d_ago	high_7d_ago	high_14d_ago	high_21d_ago	high_28d_ago	high_3d_avg	high_5d_avg	high_7d_avg	high_10d_avg	high_15d_avg	high_30d_avg	low_1d_ago	low_3d_ago	low_5d_ago	low_7d_ago	low_14d_ago	low_21d_ago	low_28d_ago	low_3d_avg	low_5d_avg	low_7d_avg	low_10d_avg	low_15d_avg	low_30d_avg	volume_1d_ago	volume_3d_ago	volume_5d_ago	volume_7d_ago	volume_14d_ago	volume_21d_ago	volume_28d_ago	volume_3d_avg	volume_5d_avg	volume_7d_avg	volume_10d_avg	volume_15d_avg	volume_30d_avg
852843	2013-01-02	25.840000	26.690001	25.780001	26.670000	21.445908	17862400.0	MDLZ	Mondelez International	Consumer Staples	Packaged Foods & Meats	Chicago, Illinois	2012-10-02	1103982	2012	NaN	NaN	NaN	NaN	NaN	NaN	0.000000	NaN	NaN	26.639999	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
852844	2013-01-03	26.700001	26.770000	26.490000	26.639999	21.421791	9075500.0	MDLZ	Mondelez International	Consumer Staples	Packaged Foods & Meats	Chicago, Illinois	2012-10-02	1103982	2012	26.670000	NaN	NaN	NaN	NaN	NaN	33.904295	NaN	NaN	26.740000	26.670000	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	21.445908	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	25.840000	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	26.690001	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	25.780001	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	17862400.0	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
852845	2013-01-04	26.700001	26.830000	26.549999	26.740000	21.502203	7696000.0	MDLZ	Mondelez International	Consumer Staples	Packaged Foods & Meats	Chicago, Illinois	2012-10-02	1103982	2012	26.653333	NaN	NaN	NaN	NaN	NaN	48.695375	NaN	NaN	26.660000	26.639999	NaN	NaN	NaN	NaN	NaN	NaN	26.683333	NaN	NaN	NaN	NaN	NaN	21.421791	NaN	NaN	NaN	NaN	NaN	NaN	21.456634	NaN	NaN	NaN	NaN	NaN	26.700001	NaN	NaN	NaN	NaN	NaN	NaN	26.413334	NaN	NaN	NaN	NaN	NaN	26.770000	NaN	NaN	NaN	NaN	NaN	NaN	26.763334	NaN	NaN	NaN	NaN	NaN	26.490000	NaN	NaN	NaN	NaN	NaN	NaN	26.273333	NaN	NaN	NaN	NaN	NaN	9075500.0	NaN	NaN	NaN	NaN	NaN	NaN	1.154463e+07	NaN	NaN	NaN	NaN	NaN
852846	2013-01-07	26.620001	26.740000	26.549999	26.660000	21.437866	7576200.0	MDLZ	Mondelez International	Consumer Staples	Packaged Foods & Meats	Chicago, Illinois	2012-10-02	1103982	2012	26.688852	NaN	NaN	NaN	NaN	NaN	39.919745	NaN	NaN	26.680000	26.740000	26.670000	NaN	NaN	NaN	NaN	NaN	26.680000	NaN	NaN	NaN	NaN	NaN	21.502203	21.445908	NaN	NaN	NaN	NaN	NaN	21.453953	NaN	NaN	NaN	NaN	NaN	26.700001	25.840000	NaN	NaN	NaN	NaN	NaN	26.673334	NaN	NaN	NaN	NaN	NaN	26.830000	26.690001	NaN	NaN	NaN	NaN	NaN	26.780000	NaN	NaN	NaN	NaN	NaN	26.549999	25.780001	NaN	NaN	NaN	NaN	NaN	26.529999	NaN	NaN	NaN	NaN	NaN	7696000.0	17862400.0	NaN	NaN	NaN	NaN	NaN	8.115900e+06	NaN	NaN	NaN	NaN	NaN
852847	2013-01-08	26.520000	26.920000	26.459999	26.680000	21.453959	14360800.0	MDLZ	Mondelez International	Consumer Staples	Packaged Foods & Meats	Chicago, Illinois	2012-10-02	1103982	2012	26.679078	NaN	NaN	NaN	NaN	NaN	55.233142	NaN	NaN	27.049999	26.660000	26.639999	NaN	NaN	NaN	NaN	NaN	26.693333	26.678	NaN	NaN	NaN	NaN	21.437866	21.421791	NaN	NaN	NaN	NaN	NaN	21.464676	21.452345	NaN	NaN	NaN	NaN	26.620001	26.700001	NaN	NaN	NaN	NaN	NaN	26.613334	26.476001	NaN	NaN	NaN	NaN	26.740000	26.770000	NaN	NaN	NaN	NaN	NaN	26.830000	26.79	NaN	NaN	NaN	NaN	26.549999	26.490000	NaN	NaN	NaN	NaN	NaN	26.519999	26.366	NaN	NaN	NaN	NaN	7576200.0	9075500.0	NaN	NaN	NaN	NaN	NaN	9.877667e+06	11314180.0	NaN	NaN	NaN	NaN

	date	open	high	low	close	adj close	volume	symbol	security	gics sector	gics sub-industry	headquarters location	date added	cik	founded	ema_9	sma_5	sma_10	sma_15	sma_30	rsi	mfi	macd	macd_signal	close_1d_next	close_1d_ago	close_3d_ago	close_5d_ago	close_7d_ago	close_14d_ago	close_21d_ago	close_28d_ago	close_3d_avg	close_5d_avg	close_7d_avg	close_10d_avg	close_15d_avg	close_30d_avg	adj close_1d_ago	adj close_3d_ago	adj close_5d_ago	adj close_7d_ago	adj close_14d_ago	adj close_21d_ago	adj close_28d_ago	adj close_3d_avg	adj close_5d_avg	adj close_7d_avg	adj close_10d_avg	adj close_15d_avg	adj close_30d_avg	open_1d_ago	open_3d_ago	open_5d_ago	open_7d_ago	open_14d_ago	open_21d_ago	open_28d_ago	open_3d_avg	open_5d_avg	open_7d_avg	open_10d_avg	open_15d_avg	open_30d_avg	high_1d_ago	high_3d_ago	high_5d_ago	high_7d_ago	high_14d_ago	high_21d_ago	high_28d_ago	high_3d_avg	high_5d_avg	high_7d_avg	high_10d_avg	high_15d_avg	high_30d_avg	low_1d_ago	low_3d_ago	low_5d_ago	low_7d_ago	low_14d_ago	low_21d_ago	low_28d_ago	low_3d_avg	low_5d_avg	low_7d_avg	low_10d_avg	low_15d_avg	low_30d_avg	volume_1d_ago	volume_3d_ago	volume_5d_ago	volume_7d_ago	volume_14d_ago	volume_21d_ago	volume_28d_ago	volume_3d_avg	volume_5d_avg	volume_7d_avg	volume_10d_avg	volume_15d_avg	volume_30d_avg
0	2013-02-20	27.070000	27.150000	26.950001	27.030001	21.735399	17057200.0	MDLZ	Mondelez International	Consumer Staples	Packaged Foods & Meats	Chicago, Illinois	2012-10-02	1103982	2012	27.307926	27.136	27.518	27.642000	27.589000	41.633625	53.176274	-0.147786	-0.050945	26.820000	26.959999	26.570000	27.680000	27.76	27.730000	28.080000	27.049999	26.903333	27.006	27.208571	27.426	27.588667	27.601333	21.679117	21.365499	22.258080	22.322405	22.298285	22.579723	21.751484	21.633545	21.716101	21.878994	22.053831	22.184635	22.194819	26.750000	26.690001	27.700001	27.799999	27.830000	27.969999	26.790001	26.886667	27.018	27.217143	27.386	27.553333	27.536667	27.190001	27.020000	27.830000	28.100000	27.980000	28.100000	27.080000	27.136667	27.248000	27.410000	27.618	27.779333	27.754667	26.750000	26.450001	27.270000	27.750000	27.67	27.820000	26.68	26.766667	26.842	27.015714	27.224	27.411333	27.382333	18297500.0	37728900.0	14931000.0	11159200.0	5800400.0	15906900.0	11671400.0	1.904973e+07	21756140.0	1.907480e+07	17005580.0	1.419575e+07	1.352419e+07
1	2013-02-21	26.990000	27.049999	26.639999	26.820000	21.566534	16936600.0	MDLZ	Mondelez International	Consumer Staples	Packaged Foods & Meats	Chicago, Illinois	2012-10-02	1103982	2012	27.252312	27.006	27.426	27.588667	27.601333	38.257648	47.431888	-0.175230	-0.078792	26.770000	27.030001	26.719999	27.750000	27.75	27.790001	27.559999	27.309999	26.936666	26.820	27.075714	27.308	27.528000	27.606000	21.735399	21.486118	22.314371	22.314371	22.346525	22.161583	21.960548	21.660350	21.566534	21.772160	21.958945	22.135851	22.198572	27.070000	26.840000	27.740000	27.730000	27.650000	27.730000	27.129999	26.936666	26.868	27.111429	27.295	27.497333	27.552333	27.150000	27.070000	27.809999	27.799999	27.950001	28.040001	27.340000	27.130000	27.096000	27.302857	27.510	27.717333	27.759000	26.950001	26.600000	27.459999	27.629999	27.65	27.299999	27.09	26.780000	26.678	26.874286	27.108	27.342667	27.388333	17057200.0	21794500.0	13902600.0	9811900.0	7541300.0	18213200.0	16348500.0	1.743043e+07	22362940.0	2.009261e+07	17608410.0	1.493817e+07	1.361005e+07
2	2013-02-22	26.889999	27.129999	26.730000	26.770000	21.526327	16664800.0	MDLZ	Mondelez International	Consumer Staples	Packaged Foods & Meats	Chicago, Illinois	2012-10-02	1103982	2012	27.165815	26.820	27.308	27.528000	27.606000	37.478423	48.958416	-0.198438	-0.104970	26.490000	26.820000	26.959999	26.570000	27.68	28.219999	27.790001	27.420000	26.873334	26.860	26.945714	27.181	27.460000	27.596667	21.566534	21.679117	21.365499	22.258080	22.692308	22.346525	22.049007	21.609420	21.598699	21.667624	21.856822	22.081171	22.191067	26.990000	26.750000	26.690001	27.700001	28.000000	27.500000	27.350000	26.983333	26.908	26.995714	27.220	27.446667	27.555667	27.049999	27.190001	27.020000	27.830000	28.320000	27.889999	27.540001	27.109999	27.118000	27.202857	27.415	27.662666	27.760667	26.639999	26.750000	26.450001	27.270000	27.93	27.350000	27.25	26.773333	26.734	26.797143	27.023	27.281333	27.390000	16936600.0	18297500.0	37728900.0	14931000.0	9623100.0	15212300.0	10162600.0	1.688620e+07	18150120.0	2.034030e+07	17828420.0	1.554640e+07	1.377650e+07
3	2013-02-25	26.790001	27.080000	26.480000	26.490000	21.301172	15527100.0	MDLZ	Mondelez International	Consumer Staples	Packaged Foods & Meats	Chicago, Illinois	2012-10-02	1103982	2012	27.086626	26.860	27.181	27.460000	27.596667	33.378362	47.675126	-0.235462	-0.132994	26.950001	26.770000	27.030001	26.719999	27.75	27.879999	27.830000	27.480000	26.693333	26.814	26.765714	27.054	27.344667	27.569333	21.526327	21.735399	21.486118	22.314371	22.418896	22.378695	22.097254	21.464678	21.561710	21.522881	21.754699	21.988429	22.169087	26.889999	27.070000	26.840000	27.740000	28.010000	27.930000	27.459999	26.890000	26.898	26.860000	27.119	27.366000	27.544333	27.129999	27.150000	27.070000	27.809999	28.150000	28.030001	27.520000	27.086666	27.120000	27.098571	27.313	27.580000	27.752000	26.730000	26.950001	26.600000	27.459999	27.83	27.639999	27.17	26.616666	26.710	26.657143	26.896	27.184667	27.369667	16664800.0	17057200.0	21794500.0	13902600.0	8954300.0	14444500.0	8688200.0	1.637617e+07	16896640.0	2.057237e+07	18265210.0	1.594000e+07	1.374912e+07
4	2013-02-26	26.530001	26.980000	26.510000	26.950001	21.671074	13702900.0	MDLZ	Mondelez International	Consumer Staples	Packaged Foods & Meats	Chicago, Illinois	2012-10-02	1103982	2012	26.967270	26.814	27.054	27.344667	27.569333	44.181951	48.178912	-0.226841	-0.152855	27.570000	26.490000	26.820000	26.959999	26.57	27.950001	27.780001	27.709999	26.736667	26.812	26.820000	26.974	27.282667	27.553667	21.301172	21.566534	21.679117	21.365499	22.475189	22.338484	22.282200	21.499524	21.560101	21.566535	21.690369	21.938574	22.156490	26.790001	26.990000	26.750000	26.690001	27.950001	27.830000	27.580000	26.736667	26.854	26.837143	26.999	27.267333	27.517000	27.080000	27.049999	27.190001	27.020000	28.110001	27.889999	27.740000	27.063333	27.077999	27.092857	27.231	27.502000	27.733333	26.480000	26.639999	26.750000	26.450001	27.85	27.690001	27.34	26.573333	26.662	26.665714	26.784	27.096667	27.345000	15527100.0	16936600.0	18297500.0	37728900.0	10961400.0	12066800.0	9863200.0	1.529827e+07	15977720.0	1.714009e+07	18654310.0	1.625657e+07	1.386713e+07

Introduction¶

Import Libraries¶

Functions¶

Read Data¶

Select Stock¶

1. Feature engineering¶

1.a. Financial indicators¶

1.b. Shift Close price in order to predict next day¶

1.c. Add lag features¶

1.d. Remove missing records after feature engineering¶

1.e. Split Data into Train ( before Covid ) and Test ( after Covid )¶

2. Train multiple regression models¶

3. Linear Regression Model¶

3.a Linear Regression Model with All Features¶

3.b. Linear Regression Model with top 20 Features¶

3.c Linear Regression Model with top 15 Features¶

3.d. Linear Regression Model with top 10 Features¶

4. Ridge Regression Parameter Fine Tuning¶

4.a. Ridge Regression with All Features¶

4.b. Ridge Regression with top 20 Features¶

5. Lasso Regression Model Parameter Fine Tuning¶

5.a Lasso Regression with All Features¶

5.b Lasso Regression with Top 20 Features¶

6. Elastic Net Regression Model Parameter Fine Tuning¶

6.a. Elastic Net with All Features¶

6.b. Elastic Net with Top 20 Features¶

7. Model Comparison¶

7.a Final Best Model¶

8. Train Multiple Different Stocks with Ridge Regression Model¶

	Model	Mean Squared Error	Mean Absolute Error	R2 Score	Training Time (s)
0	Ridge Regression	0.746614	0.598242	0.98175	0.002745
1	Linear Regression	0.77558	0.613905	0.981042	0.015371
2	Lasso Regression	1.066347	0.782536	0.973934	0.104589
3	Elastic Net	1.070319	0.718474	0.973837	0.108071
4	Gradient Boosting	87.489334	7.501632	-1.138585	2.444588
5	AdaBoost	88.43475	7.561282	-1.161695	0.786036
6	Random Forest	90.124875	7.670387	-1.203008	4.302596
7	Decision Tree	91.634349	7.746372	-1.239906	0.079427
11	CatBoost	93.180853	7.853729	-1.277709	4.704044
8	XGBoost	94.682514	7.893578	-1.314415	0.665853
9	K-Neighbors Regressor	330.959365	16.750109	-7.089956	0.000767
10	SVR	349.552891	17.672316	-7.544455	0.150409

	date	y_test	lr_pred_base
1729	2020-01-02	54.240002	54.157799
1730	2020-01-03	54.150002	54.553547
1731	2020-01-06	53.919998	54.336899
1732	2020-01-07	54.049999	53.907121
1733	2020-01-08	54.189999	54.192608

	Feature	Importance
0	adj close_10d_avg	59.927343
1	close_10d_avg	52.842967
2	close_15d_avg	32.217386
3	ema_9	31.326586
4	adj close_15d_avg	29.982687
5	close_5d_avg	17.929065
6	sma_5	13.495880
7	adj close_5d_avg	12.461717
8	adj close_3d_avg	10.395039
9	close_1d_ago	9.861051
10	adj close_7d_ago	9.733592
11	adj close_1d_ago	9.044414
12	close_7d_ago	7.605479
13	sma_15	6.911237
14	adj close_14d_ago	5.986732

	Feature	Importance
0	adj close_10d_avg	39.701594
1	close_10d_avg	34.492844
2	ema_9	16.380613
3	adj close_7d_avg	14.543602
4	sma_5	11.032622
5	close_15d_avg	10.687774
6	sma_15	9.954755
7	adj close_5d_avg	8.830203
8	adj close_7d_ago	8.207036
9	adj close_15d_avg	7.303476
10	close_7d_ago	7.068989
11	adj close_1d_ago	6.587839
12	close_5d_avg	6.427914
13	close_1d_ago	4.837201
14	adj close_3d_avg	4.708264
15	close_3d_ago	4.101885
16	adj close_3d_ago	3.394063
17	low_5d_avg	1.134743
18	low_10d_avg	0.507845
19	adj close_14d_ago	0.220884

	Feature	Importance
0	close_10d_avg	29.471317
1	adj close_10d_avg	24.455782
2	ema_9	21.480216
3	adj close_5d_avg	16.328249
4	adj close_1d_ago	10.320838
5	adj close_3d_avg	9.162211
6	close_1d_ago	7.864891
7	adj close_7d_ago	7.162200
8	sma_15	6.734176
9	close_7d_ago	6.356886
10	close_5d_avg	4.707638
11	close_15d_avg	3.343602
12	sma_5	2.379653
13	adj close_15d_avg	2.216233
14	adj close_14d_ago	0.415949

	Feature	Importance
0	ema_9	16.842727
1	close_5d_avg	14.158982
2	adj close_3d_avg	8.615665
3	adj close_5d_avg	6.493537
4	close_15d_avg	6.467632
5	close_10d_avg	6.069412
6	sma_5	3.130954
7	adj close_10d_avg	1.666197
8	close_1d_ago	0.905094
9	adj close_15d_avg	0.029654

	Feature	Importance
0	close_5d_avg	19.627386
1	ema_9	18.897025
2	sma_5	18.403314
3	adj close_5d_avg	11.844801
4	close_10d_avg	9.872194
5	adj close_10d_avg	9.389002
6	close_15d_avg	7.887685
7	close_1d_ago	7.038813
8	close_7d_avg	6.958126
9	adj close_1d_ago	6.600528
10	close_5d_ago	6.008206
11	sma_15	5.538994
12	low_5d_avg	4.965283
13	low_10d_avg	4.840114
14	open_5d_avg	4.446916
15	high_5d_avg	3.714418
16	open_10d_avg	3.486739
17	high_30d_avg	2.896473
18	close_3d_ago	2.893431
19	open_15d_avg	2.892245

	Feature	Importance
0	sma_5	25.441510
1	close_5d_avg	22.435160
2	adj close_5d_avg	11.490239
3	adj close_1d_ago	5.671591
4	adj close_10d_avg	5.507894
5	close_5d_ago	5.147037
6	close_1d_ago	4.554521
7	close_10d_avg	4.509604
8	close_15d_avg	3.719596
9	low_5d_avg	3.022988
10	sma_15	2.915010
11	ema_9	2.899160
12	low_10d_avg	1.985008
13	high_5d_avg	1.745675
14	open_10d_avg	1.590243
15	open_5d_avg	1.561036
16	open_15d_avg	1.264441
17	close_3d_ago	0.798379
18	close_7d_avg	0.640924
19	high_30d_avg	0.200250

	Feature	Importance
0	ema_9	4.177865
1	close_3d_avg	1.314731
2	macd	1.312350
3	macd_signal	1.160117
4	sma_15	1.006266
5	adj close_3d_avg	0.822478
6	low_1d_ago	0.722646
7	close_3d_ago	0.583764
8	rsi	0.543082
9	open_3d_ago	0.510853
10	sma_30	0.492814
11	high_1d_ago	0.466770
12	adj close_3d_ago	0.422750
13	low_15d_avg	0.404237
14	open_1d_ago	0.365671
15	high_14d_ago	0.307277
16	low_30d_avg	0.291303
17	sma_10	0.255037
18	open_30d_avg	0.241245
19	high_30d_avg	0.227913

	Feature	Importance
0	ema_9	4.017287
1	close_3d_avg	2.790781
2	low_1d_ago	1.047406
3	close_3d_ago	1.034968
4	sma_30	1.029511
5	low_15d_avg	0.997798
6	macd	0.968169
7	macd_signal	0.880009
8	high_14d_ago	0.568316
9	high_1d_ago	0.533765
10	rsi	0.496376
11	sma_10	0.423288
12	adj close_3d_avg	0.228550
13	open_3d_ago	0.204478
14	open_1d_ago	0.139447
15	low_30d_avg	0.025033
16	sma_15	0.000000
17	adj close_3d_ago	0.000000
18	open_30d_avg	0.000000
19	high_30d_avg	0.000000

	date	y_test	lr_pred_base	lr_pred20	lr_pred15	lr_pred10	ridge_pred_base	ridge_pred20	lasso_pred_base
1729	2020-01-02	54.240002	54.157799	54.239022	54.554907	54.590333	54.205840	54.349344	54.496024
1730	2020-01-03	54.150002	54.553547	54.520826	54.558027	54.511431	54.542331	54.381596	54.190970
1731	2020-01-06	53.919998	54.336899	54.065422	54.148986	54.126188	54.345656	54.233187	53.975755
1732	2020-01-07	54.049999	53.907121	54.067429	53.903359	53.992480	53.873857	54.036503	53.828707
1733	2020-01-08	54.189999	54.192608	53.987340	53.942897	54.081508	54.195624	54.059649	54.002350

	date	y_test	lr_pred_base	lr_pred20	lr_pred15	lr_pred10	ridge_pred_base	ridge_pred20	lasso_pred_base	lasso_pred20
1729	2020-01-02	54.240002	54.157799	54.239022	54.554907	54.590333	54.205840	54.349344	54.496024	54.501734
1730	2020-01-03	54.150002	54.553547	54.520826	54.558027	54.511431	54.542331	54.381596	54.190970	54.262946
1731	2020-01-06	53.919998	54.336899	54.065422	54.148986	54.126188	54.345656	54.233187	53.975755	54.001497
1732	2020-01-07	54.049999	53.907121	54.067429	53.903359	53.992480	53.873857	54.036503	53.828707	53.859163
1733	2020-01-08	54.189999	54.192608	53.987340	53.942897	54.081508	54.195624	54.059649	54.002350	53.989285

	date	y_test	lr_pred_base	lr_pred20	lr_pred15	lr_pred10	ridge_pred_base	ridge_pred20	lasso_pred_base	lasso_pred20	elastic_pred_base
1729	2020-01-02	54.240002	54.157799	54.239022	54.554907	54.590333	54.205840	54.349344	54.496024	54.501734	54.444972
1730	2020-01-03	54.150002	54.553547	54.520826	54.558027	54.511431	54.542331	54.381596	54.190970	54.262946	54.198628
1731	2020-01-06	53.919998	54.336899	54.065422	54.148986	54.126188	54.345656	54.233187	53.975755	54.001497	54.011290
1732	2020-01-07	54.049999	53.907121	54.067429	53.903359	53.992480	53.873857	54.036503	53.828707	53.859163	53.818335
1733	2020-01-08	54.189999	54.192608	53.987340	53.942897	54.081508	54.195624	54.059649	54.002350	53.989285	54.037049

	date	y_test	lr_pred_base	lr_pred20	lr_pred15	lr_pred10	ridge_pred_base	ridge_pred20	lasso_pred_base	lasso_pred20	elastic_pred_base	elastic_pred20
1729	2020-01-02	54.240002	54.157799	54.239022	54.554907	54.590333	54.205840	54.349344	54.496024	54.501734	54.444972	54.503795
1730	2020-01-03	54.150002	54.553547	54.520826	54.558027	54.511431	54.542331	54.381596	54.190970	54.262946	54.198628	54.263269
1731	2020-01-06	53.919998	54.336899	54.065422	54.148986	54.126188	54.345656	54.233187	53.975755	54.001497	54.011290	54.055709
1732	2020-01-07	54.049999	53.907121	54.067429	53.903359	53.992480	53.873857	54.036503	53.828707	53.859163	53.818335	53.920259
1733	2020-01-08	54.189999	54.192608	53.987340	53.942897	54.081508	54.195624	54.059649	54.002350	53.989285	54.037049	54.058664

	Feature	Importance
0	ema_9	1.613307
1	sma_5	1.518359
2	macd	1.493389
3	close_3d_avg	1.290926
4	sma_10	1.289912
5	macd_signal	1.253588
6	adj close_3d_avg	1.239816
7	low_1d_ago	0.982940
8	sma_15	0.923173
9	high_1d_ago	0.815800
10	close_3d_ago	0.765736
11	open_1d_ago	0.631358
12	sma_30	0.586684
13	low_3d_avg	0.580421
14	adj close_3d_ago	0.544949
15	open_3d_ago	0.531747
16	rsi	0.485403
17	high_3d_avg	0.431768
18	close_1d_ago	0.425664
19	low_15d_avg	0.399066

	Feature	Importance
0	ema_9	3.518012
1	close_3d_avg	2.975563
2	low_15d_avg	1.164547
3	sma_30	1.008792
4	low_1d_ago	0.912273
5	high_1d_ago	0.840268
6	sma_10	0.826143
7	adj close_3d_avg	0.783827
8	macd_signal	0.701467
9	macd	0.690383
10	sma_15	0.565839
11	adj close_3d_ago	0.556647
12	rsi	0.506954
13	close_3d_ago	0.408292
14	open_3d_ago	0.238611
15	high_3d_avg	0.039417
16	low_3d_avg	0.000000
17	sma_5	0.000000
18	close_1d_ago	0.000000
19	open_1d_ago	0.000000

	MSE	RMSE	MAE	R2	Model
0	0.729511	0.854114	0.585601	0.982168	Ridge with Top 20 Features
1	0.758611	0.870983	0.605987	0.981457	Ridge with All Features
2	0.768289	0.876521	0.613346	0.98122	Linear Reg. with Top 20 Features
3	0.77558	0.88067	0.613905	0.981042	Linear Reg. with All Features
4	0.786499	0.886848	0.626022	0.980775	Linear Reg. with Top 10 Features
5	0.796893	0.892688	0.626154	0.980521	Linear Reg. with Top 15 Features
6	0.90796	0.952869	0.652726	0.977806	Elastic_Net with All Features
7	0.940304	0.969693	0.663218	0.977015	Lasso with All Features
8	0.948047	0.973677	0.664517	0.976826	Lasso with Top 20 Features
9	0.948242	0.973777	0.666597	0.976821	Elastic_Net with Top 20 Features

	symbol	min_date	max_date	days_cnt	min_year	max_year
0	A	2013-01-02	2023-11-08	2733	2013	2023
1	AAL	2013-01-02	2023-11-08	2733	2013	2023
2	AAPL	2013-01-02	2023-11-08	2733	2013	2023
3	ABBV	2013-01-02	2023-11-08	2733	2013	2023
4	ABNB	2020-12-10	2023-11-08	733	2020	2023

	symbol	security	gics sector	volume
0	TSLA	Tesla, Inc.	Consumer Discretionary	3.009291e+10
1	AMD	AMD	Information Technology	1.342035e+10
2	AMZN	Amazon	Consumer Discretionary	1.305160e+10
3	AAPL	Apple Inc.	Information Technology	1.303964e+10
4	F	Ford Motor Company	Consumer Discretionary	1.278319e+10

	gics sector	volume
0	Consumer Discretionary	9.171407e+10
1	Information Technology	8.888840e+10
2	Financials	6.728113e+10
3	Communication Services	5.267892e+10
4	Health Care	3.755560e+10
5	Industrials	3.672492e+10
6	Energy	3.245171e+10
7	Consumer Staples	2.824873e+10
8	Utilities	2.214882e+10
9	Materials	1.432867e+10
10	Real Estate	1.318748e+10

	date	open	high	low	close	adj close	volume	symbol	security	gics sector	gics sub-industry	headquarters location	date added	cik	founded
0	2013-01-02	18.003504	18.193193	17.931683	18.099348	18.099348	101550348.0	GOOGL	Alphabet Inc. (Class A)	Communication Services	Interactive Media & Services	Mountain View, California	2014-04-03	1652044	1998
1	2013-01-03	18.141392	18.316566	18.036036	18.109859	18.109859	92635272.0	GOOGL	Alphabet Inc. (Class A)	Communication Services	Interactive Media & Services	Mountain View, California	2014-04-03	1652044	1998
2	2013-01-04	18.251753	18.555305	18.210211	18.467718	18.467718	110429460.0	GOOGL	Alphabet Inc. (Class A)	Communication Services	Interactive Media & Services	Mountain View, California	2014-04-03	1652044	1998
3	2013-01-07	18.404655	18.503002	18.282784	18.387136	18.387136	66161772.0	GOOGL	Alphabet Inc. (Class A)	Communication Services	Interactive Media & Services	Mountain View, California	2014-04-03	1652044	1998
4	2013-01-08	18.406906	18.425926	18.128880	18.350851	18.350851	66976956.0	GOOGL	Alphabet Inc. (Class A)	Communication Services	Interactive Media & Services	Mountain View, California	2014-04-03	1652044	1998

	MSE	RMSE	MAE	R2	symbol
0	56.919689	7.544514	5.112978	0.995061	NVDA
1	0.559182	0.747785	0.529467	0.993251	VZ
2	7.770865	2.787627	2.107652	0.992581	AAPL
3	5.039335	2.244846	1.670374	0.992108	GOOG
4	4.98575	2.232879	1.654225	0.992002	GOOGL
5	46.381058	6.810364	4.690443	0.990596	META
6	2.057386	1.434359	1.020137	0.990386	CVS
7	0.147953	0.384647	0.271314	0.990362	F
8	1.27562	1.129434	0.839937	0.98989	GM
9	28.055837	5.296776	3.992424	0.989314	MSFT
10	84.835254	9.210606	6.497099	0.988764	TSLA
11	0.575533	0.758639	0.55086	0.988709	PFE
12	0.581909	0.76283	0.577919	0.988411	BAC
13	0.26222	0.512075	0.377068	0.988041	KEY
14	1.80572	1.343771	0.958788	0.987257	INTC
15	0.991949	0.995966	0.734726	0.987031	WFC
16	11.147927	3.338851	2.469883	0.985621	AMZN
17	0.174755	0.418037	0.281833	0.983153	T
18	10.145635	3.185221	2.328928	0.982675	AMD
19	2.409729	1.55233	1.141577	0.981822	C
20	0.124353	0.352637	0.259468	0.979688	HBAN
21	0.870412	0.932958	0.684029	0.978487	BMY
22	2.087087	1.444675	1.154954	0.96863	CCL
23	4.398395	2.097235	1.52279	0.967914	JNJ
24	4.070663	2.017589	1.528408	0.589411	VTRS